Sectors

Highlights The German 10-year bund yield rising to 1%, or the U.S. 10-year T-bond yield rising to 3% would be a trigger to downgrade equities and upgrade bonds... ...especially as the blue sky expectations for global growth in H1 2018 will turn out to be overly-optimistic. On a 6-9 month horizon, upgrade Airlines to overweight. Downgrade Banks to underweight. Upgrade Germany (DAX) to neutral. Downgrade Italy (MIB) and Spain (IBEX) to underweight. Feature Where has the equity market cycle gone? Since 2012, the stock market's 6-month returns have generated an unprecedented consistency, with only a brief breakdown - at the end of 2015 - into negative territory (Chart of the Wesk and Chart I-2). Chart of the WeekSince 2012, The Equity Market ##br##Cycle Has Disappeared Chart I-2Much Less Cyclicality In Equities ##br##Than In Commodities The disappearance of the equity market cycle brings to mind the concept of the "Great Moderation", a term coined in 2002 to describe the big drop in business cycle volatility during the 1990s. In 2004, Ben Bernanke suggested that "improvements in monetary policy, though certainly not the only factor, probably were an important source of the Great Moderation." Today's Great Moderation 2.0 refers to the equity market cycle - or rather, its disappearance. And in finding a reason for the Great Moderation 2.0, Bernanke's attribution to monetary policy might be right on the money. Stick With TINA, Or Flirt With TIA? For many years, ultra-accommodative monetary policy has provided a consistent and substantial uplift to world stock market valuations. Since 2012, our preferred measure of equity market valuation - world stock market capitalisation to GDP - has almost doubled. This inexorable and relatively trouble-free rise has even spawned its own acronym: TINA - There Is No Alternative (to owning equities.) However, the uplift to stock market valuations has happened in a less obvious way than you might realise. Based on the excellent predictive power of stock market capitalisation to GDP, the prospective 10-year annualised return from world equities has collapsed from 9% in 2012 to 1.5% now (Chart I-3). Over the same period, the global 10-year bond yield has compressed from 3% to 1.5%. Hence, the collapse in prospective equity returns is not due to the decline in bond yields per se. It has happened mostly because the excess return offered by equities over bonds - the so-called 'equity risk premium' has compressed from 6% to zero (Chart I-4). Chart I-3World Equity Market Cap To GDP Implies##br## A Feeble Prospective 10-Year Return Chart I-4Prospective Equity Returns ##br##Have Become 'Bond Like' Ultra-accommodative monetary policy has caused the disappearance of the equity risk premium. The simple reason is that at low bond yields, the risk of owning bonds becomes similar to the risk of owning equities. Chart I-5Below A 2% Yield, 10-Year Bonds Have ##br##More Negative Skew Than Equities When bond yields approach their lower bound, bond prices have little upside but they have a lot of downside. This ratio of an investment's potential losses relative to its potential gains is the risk that most frightens investors,1 and is known as negative skew. At yields below 2%, bond returns become as negatively skewed as equity returns, or even more negatively skewed than equities (Chart I-5). As the risk of bonds increases to become 'equity-like', the prospective return from equities must compress to become 'bond-like'. Which is to say, equity valuations become substantially richer. All well and good - so long as the global 10-year bond yield stays low. Above a 2% yield, the negative skew on bond returns disappears, and equities once again require an excess prospective return over bonds. More colloquially, investors would dump TINA and start flirting with TIA (There Is an Alternative). In essence, a big threat to the Great Moderation 2.0 comes the global 10-year bond yield rising to 2% - broadly equivalent to the German 10-year bund yield rising to 1%, or the U.S. 10-year T-bond yield rising to 3%. Any moves towards these thresholds would be a trigger to downgrade equities and upgrade bonds - especially as we now explain why the blue sky expectations for global growth in H1 2018 will turn out to be overly-optimistic. The Equity Sector Cycle Is Alive And Well For the stock market in aggregate, the cycle has been moribund. But for equity sector relative performance, the cycle is very much alive and well. In The Cobweb Theory And Market Cycles 2 we showed and explained the existence of mini-cycles in economic and financial variables. To summarise, a lag between the demand for credit and its supply necessarily creates mini-cycles in both the price of credit (the bond yield) and the quantity of credit (the global credit impulse). Thereby it also creates mini-cycles in GDP growth. The useful point is that these cycles are very regular with half-cycles averaging 6-8 months. Which makes their turning points and phases predictable. Given that the global credit impulse cycle has been in a mini-upswing phase since last May, it is highly likely to turn into a mini-downswing phase through the first half of 2018. The latest data point, showing a tick down, seems to corroborate such a turning point. From an equity sector perspective, Banks versus Healthcare has closely tracked the phases of the credit impulse mini-cycle (Chart I-6). In all five of the last five mini-downswings, Banks have underperformed Healthcare, and we would expect no difference in the next mini-downswing. Hence, on a 6-9 month horizon, downgrade Banks to underweight. Unsurprisingly, exactly the same pattern applies to Basic Materials (and Energy) versus Healthcare (Chart I-7). Hence, on a 6-9 month horizon, stay underweight Basic Materials and Energy versus Healthcare. Also unsurprisingly, the performance of European Airlines is a mirror-image of the oil price cycle, given that aviation fuel comprises the sector's main variable cost (Chart I-8). As an aside, this also somewhat insulates the European Airlines against a strengthening euro, given that this variable cost is priced in dollars. Hence, on a 6-9 month horizon, upgrade European Airlines to overweight. Chart I-6Banks Vs. Healthcare Tracks The ##br##Credit Impulse Mini-Cycle Chart I-7Materials Vs. Healthcare Tracks The##br## Credit Impulse Mini-Cycle Chart I-8European Airlines Relative Performance Is A##br## Mirror-Image of The Oil Price Cycle Country Allocation Just Drops Out Of Sector Allocation Our core philosophy of investment reductionism teaches us that for most stock markets, the sector (and dominant company) skews swamp any effect that comes from the domestic economy. For example, the defining skew for Italy's MIB and Spain's IBEX is their large overweighting to banks. So unsurprisingly, MIB and IBEX relative performance reduces to: will banks outperform the market? (Chart I-9 and Chart I-10). Chart I-9Italy = Long Banks Chart I-10Spain = Long Banks Therefore, the key consideration for European equity country allocation is always: how to allocate to the vital few equity sectors that feature most often in the skews: Banks, Healthcare, Energy and Materials. To reiterate, our 6-9 month recommendation is to underweight Banks, Materials And Energy versus Healthcare, and to overweight Airlines versus the market. Then to arrive at a country allocation, combine the cyclical view on the vital few sectors with the country sector skews shown in Box I-1. Even if you disagree with our sector views, the sector-based approach is the right way to pick European equity markets. If you agree with our sector views, the result is the following updated European equity market allocation: Box I-1: The Vital Few Sector Skews That Drive Country Relative Performance For major equity indexes in the euro area, the dominant sector skews that drive relative performance are as follows: Germany (DAX) is overweight Chemicals, underweight Banks. France (CAC) is underweight Banks and Basic Materials. Italy (MIB) is overweight Banks. Spain (IBEX) is overweight Banks. Netherlands (AEX) is overweight Technology, underweight Banks. Ireland (ISEQ) is overweight Airlines (Ryanair) which is, in effect, underweight Energy. And for major equity indexes outside the euro area: The U.K. (FTSE100) is effectively underweight the pound. Switzerland (SMI) is overweight Healthcare, underweight Energy. Sweden (OMX) is overweight Industrials. Denmark (OMX20) is overweight Healthcare and Industrials. Norway (OBX) is overweight Energy. The U.S. (S&P500) is overweight Technology, underweight Banks. Overweight: France, Ireland, U.K., Switzerland and Denmark. Neutral: Germany, Netherlands. Underweight: Italy, Spain, Sweden and Norway. In terms of change, it means upgrading Germany (DAX) to neutral and downgrading Italy (MIB) and Spain (IBEX) to underweight. Dhaval Joshi, Senior Vice President Chief European Investment Strategist dhaval@bcaresearch.com 1 Please see the European Investment Strategy Weekly Report "Are Bonds A Greater Risk Than Equities", January 28, 2018 available at eis.bcaresearch.com. 2 Please see the European Investment Strategy Weekly Report "The Cobweb Theory And Market Cycles", January 11, 2018 available at eis.bcaresearch.com. Fractal Trading Model* There is a lot of optimism already priced into the South African rand, making it vulnerable to a countertrend reversal. Therefore, this week's recommended trade is to go long USD/ZAR with a profit-target of 6% and a symmetrical stop-loss. In other trades, short S&P500/long Eurostoxx50 hit its stop-loss, while short Japanese energy and short palladium moved comfortably into profit. For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-11 The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields Interest Rate Chart II-5Indicators To Watch ##br##- Interest Rate Expectations Chart II-6Indicators To Watch ##br##- Interest Rate Expectations Chart II-7Indicators To Watch ##br##- Interest Rate Expectations Chart II-8Indicators To Watch ##br##- Interest Rate Expectations

Neutral Prior to our midsummer upgrade to neutral, our key concern for the S&P chemicals index was that perennial overcapacity would put a ceiling on margin improvement. Since then surging global demand growth has driven producer prices sky high (second panel), in line with our expectations. More surprising is the newly-found industry discipline which has thus far refrained from splurging on new capacity (third panel), likely assisted by a wave of consolidation in the space. The upshot is that our productivity proxy has perked up to its highest level in nearly a decade, as have sell-side earnings expectations. While it remains too early for us to turn bullish, the outlook has certainly brightened for chemicals; maintain a neutral stance. The ticker symbols for the stocks in this index are: BLBG: S5CHEM - APD, ARG, CF, DOW, EMN, ECL, DD, FMC, IFF, LYB, MON, MOS, PPG, PX, SHW.

Neutral In yesterday’s Weekly Report, we outlined that the most cyclical parts of the S&P industrials index with high foreign sales content would benefit disproportionately from our stable-to-mildly sanguine EM/China view. While the broad machinery index fits the bill, the industrial machinery sub index less so, and we recommend monetizing gains of 4% since inception and moving to the sidelines while redeploying profits into the more cyclical S&P construction machinery & heavy truck index. One key determinant of the relative move of these indexes is the U.S. dollar. The greenback troughed in 2011 and since then the more “defensive”, less globally-exposed S&P industrials machinery index left their brethren in the dust (top panel). Now that the U.S. dollar has peaked, the catch up phase in the S&P construction machinery & heavy truck index that is already underway will likely gain momentum (bottom panel). Bottom Line: Book profits of 4% in the S&P industrial machinery index and downgrade to a benchmark allocation while staying overweight construction machinery; please see yesterday’s Weekly Report for more details. The ticker symbols for the stocks in this index are: BLBG: S5INDM - ITW, IR, SWK, PH, FTV, DOV, PNR, XYL, SNA, FLS.

Highlights Portfolio Strategy A stable China, a depreciating U.S. dollar, rising commodity prices and sustained synchronized global growth signal that the industrials complex, especially the most cyclical part, remains on a solid footing. Deteriorating profit prospects warn that investors should refrain from paying a premium valuation for industrial machinery; take profits and move to the sidelines. Recent Changes S&P Industrial Machinery - Book profits of 4% and downgrade to neutral today. S&P Construction Machinery & Heavy Truck - Stop triggered last week, remove from the high-conviction list for a 10% gain. Small Caps / Large Caps - Downgrade alert in a recent Insight. Table 1 Feature The S&P 500 smashed through the 2,800 mark last week, as corporate profits continued to deliver, the U.S. dollar took a dive and global economic data releases held their own. Stars could not be more aligned for a euphoric blow off phase, with equity bourses the world over already registering annual-like returns in but a few short weeks. While stocks have more room to run, especially versus bonds, on a cyclical time frame, tactically the likelihood of a short-term healthy pullback is increasing. Last week we identified five indicators we are closely monitoring that are signaling an overstretched market.1 This week we update our Complacency-Anxiety Indicator that also catapulted to all-time highs and breached the one standard deviation above the historical mean mark (Chart 1). This confirms that a Q1 setback remains likely, and our strategy since December 18 has been to monetize gains in tactical trades and institute stops to the high flyers in our high-conviction call list. Were a 5-10% correction to materialize, we would "buy the dip" as we do not foresee a recession in the coming 9-12 months. While consumer price inflation is nowhere to be found, corporate selling prices are climbing at a brisk pace. The U.S. dollar debasement and related commodity reflex rebound, especially in oil prices, are the culprits, and the latter will likely assist even the CPI basket and morph into an inflationary impulse as we posited in late-November (please see the bottom two panels of Chart 1B). Already, inflation expectations are headed higher. Chart 2 updates our corporate sector pricing power proxy and our diffusion index. It also updates the business sector's overall wage inflation and associated diffusion index from the latest BLS employment report. The middle panel of Chart 2 shows the Atlanta Fed Wage Growth Tracker and that measure of wage inflation has converged down to the AHE reading, suffering a 100bps drop in the past year. Chart 1Complacency Reigns Chart 2Margin Expansion Phase Is Intact Corporate pricing power is upbeat at a time when wages are decelerating. Taken together, our margin proxy indicator suggests that the ongoing profit margin expansion phase has more upside (bottom panel, Chart 2). Table 2 shows our updated industry group pricing power gauges, which we calculate from the relevant CPI, PPI, PCE and commodity growth rates for each of the 60 industry groups we track. Table 2 also highlights shorter-term pricing power trends and each industry's spread to overall inflation. Table 2Industry Group Pricing Power 78% of the industries we cover are lifting selling prices, and 45% are doing so at a faster clip than overall inflation. Importantly, inflation rates have increased since our late-September update. The outright deflating sectors dropped by two to 13 since our last update. Encouragingly, only 7 industries are experiencing a downtrend in selling price inflation, or 5 fewer than our most recent report. Impressively, deep cyclicals/commodity-related industries dominate the top ranks, occupying 8 out of the top 10 slots (top panel, Chart 3). A softening greenback and rising global end demand explain the commodity complex's sustained ability to increase prices. In contrast, tech, telecom and consumer discretionary sectors populate the bottom ranks of Table 2. Netting it out, accelerating corporate sector pricing power will continue to bolster top line growth in 2018. Tack on high operating leverage kicking into higher gear at this stage of the cycle and still muted wage inflation and profit margins and EPS growth will remain upbeat. With regard to cyclicals versus defensives, diverging pricing power (Chart 3) and wage growth trends (Chart 4) suggest that cyclicals continue to have the upper hand compared with defensives (Chart 5). Chart 3Deep Cyclicals... Chart 4...Have The Upper Hand... Chart 5...Vs. Defensives This week we update our view on a deep cyclical sector and modestly tweak our intra-sector positioning. Industrials And China We lifted the S&P industrials sector to an above benchmark allocation in early October via boosting the S&P construction machinery & heavy truck sub index to overweight.2 Synchronized global growth, a capex upcycle, firming capital goods final demand, and the U.S. dollar's fall coupled with the commodity price rebound all pointed to a bright outlook for U.S. capital goods producers. Currently, all these forces remain in play and continue to bolster industrials stocks' profit prospects. However, the emerging market (EM)/Chinese economic backdrop deserves closer scrutiny. Why? Because the most cyclical parts of the industrials complex are levered to the EM in general and China in particular. These high operating leverage businesses also drive relative profit and stock performance, signaling that China's economic growth might or ails determine the overall fortunes of U.S. capital goods producers. While Chinese economic data are currently a mixed bag and we take them with a big grain of salt, global high-frequency financial market data are emitting an unambiguously positive signal. First, BCA's FX strategist, Mathieu Savary, brought to our attention that the extremely economic-sensitive Canadian TSX Venture Exchange Index is in a V-shaped recovery.3 Highly speculative basic resources issues dominate this Index and help explain the tight positive correlation with Chinese output (top panel, Chart 6). Second, the ultimate economic-sensitive indicator, Dr. Copper, is also in a violent upswing, heralding that China will be, at least, stable in 2018 (middle panel, Chart 6). Third, high-beta Australian materials stocks have been in an upward trajectory since the early 2016 trough both versus the MSCI All-Country World Index and the broad Australian market, sniffing out improving Chinese-related commodity demand (bottom panel, Chart 6). Similarly, upbeat non-Chinese economic data suggest that China's economic prospects are far from faltering. Australia's close economic ties with China signal that taking a pulse of the Australian economic juggernaut reveals the state of China's economic affairs. Down Under employment growth has been brisk of late, with annual job creation running at a 3.3% clip, a rate last hit in the mid-2000s when China's economy was roaring and the commodity super-cycle was in full swing (second panel, Chart 7). Australian CEO confidence as well as consumer confidence are pushing decade highs, and the manufacturing PMI survey recently shot to a 16 year high (third panel, Chart 7). Chart 6China Is##BR##Alright Chart 7Australian Indicators Confirm:##BR## China Is Stable All of this suggests that China will likely remain stable in 2018, barring a policy mistake a la the August 11, 2015 currency devaluation. The upshot is that industrials EPS and equities have more room to run. On that front, both our Cyclical Macro Indicator and our profit growth model corroborate that the path of least resistance for relative share prices is higher (Chart 8). U.S. dollar debasing is synonymous with capital goods producers' top line growth acceleration, as a large part of total revenues are sourced from abroad. The near 20 percentage point fall in the trade-weighted U.S. dollar since 2015 suggests that more global market share gains are in store for U.S. industrials (Chart 9). Global growth is also joined at the hip with the greenback's depreciation. Synchronized global growth along with our derivative coordinated global capex growth 2018 theme, will likely serve as catalysts for a sustained breakout in relative share prices (Chart 10). Chart 8EPS Model And CMI Flash Green Chart 9Industrials Love A Cheap Greenback Chart 10Levered To Global Growth Adding it up, a stable China is music to the ears of industrials executives. Tack on a depreciating U.S. dollar, rising commodity prices and sustained synchronized global growth and the most cyclical parts of the industrials complex will continue to lead the pack. Bottom Line: Stay overweight the S&P industrials index, but selectivity is warranted. Take Profits In Industrial Machinery We outlined above that the most cyclical parts of the S&P industrials index with high foreign sales content would benefit disproportionately from our stable-to-mildly sanguine EM/China view. While the broad machinery index fits the bill, the industrial machinery sub index less so, and we recommend monetizing gains of 4% since inception and moving to the sidelines. Chart 11 shows the relative performance of the two key drivers of the S&P machinery index: industrial machinery and construction machinery & heavy truck sub-indexes. While these indexes moved hand-in-hand since the mid-1990s, early this decade this tight positive correlation fell apart. One key determinant of the relative move of these indexes is the U.S. dollar. The greenback troughed in 2011 and since then the more "defensive", less globally-exposed S&P industrials machinery index left their brethren in the dust (bottom panel, Chart 11). Now that the U.S. dollar has peaked, the catch up phase in the S&P construction machinery & heavy truck index that is already underway will likely gain momentum (top panel, Chart 11). Beyond the depreciating currency, at the margin, softening S&P industrial machinery operating metrics argue for pruning exposure in this index. Both the Empire and Philly Fed new orders surveys have petered out, suggesting that industry new order growth will likely continue to lose steam (middle panel, Chart 12). In fact, a weak industrial machinery new orders-to-inventories ratio is also warning that sell-side analysts' relative profits forecasts are too optimistic (bottom panel, Chart 12). Chart 11Catch Up Phase Chart 12Waning End-Demand Drilling deeper into industry operating metrics is revealing. While shipments have held their own and moved mostly sideways similar to new orders, inventory accumulation is worrying. Industry inventories have risen by over 30% during the past three years (Chart 13). Simultaneously, industrial machinery backlogs have drifted steadily lower. Given the supply build up, any hiccup in demand, even a minor one, could prove very deflationary and heavily weigh on industry profitability. With regard to valuations, Chart 14 shows that both on a relative trailing price-to-sales and relative forward price-to-earnings ratio basis, the index is trading one standard deviation above the historical mean. The moderating industry demand backdrop suggests that relative valuations are expensive. Chart 13Inventory Liquidation Risk Chart 14Why Pay A Premium? Adding it all up, deteriorating profit prospects warn that investors should refrain from paying a premium valuation for the S&P industrial machinery index. Bottom Line: Book profits of 4% in the S&P industrial machinery index and downgrade to a benchmark allocation. We also recommend redeploying profits from our downgrade in the S&P industrial machinery index to their more cyclical machinery siblings the S&P construction machinery & heavy truck index, thus sustaining the overall overweight exposure in the broad S&P industrials sector. Housekeeping Last week we instituted a risk management tool for our 2018 high-conviction list: setting a stop once a call has cleared the 10% return mark.4 This past week, the S&P construction machinery & heavy truck index hit the trailing stop at the 10% mark, and thus we are booking gains and removing this index from the high-conviction list. While our confidence is not as high as in late-November given the parabolic move in this index and rising chance of a tactical overall equity market pullback, from a cyclical perspective we continue to recommend a core overweight in this industrials sector powerhouse. Anastasios Avgeriou, Vice President U.S. Equity Strategy anastasios@bcaresearch.com 1 Please see BCA U.S. Equity Strategy Weekly Report, "Too Good To Be True?" dated January 22, 2018, available at uses.bcaresearch.com. 2 Please see BCA U.S. Equity Strategy Weekly Report, "Earnings Take Center Stage," dated October 2, 2017, available at uses.bcaresearch.com. 3 Please see BCA Foreign Exchange Strategy Weekly Report, "Health Care Or Not, Risks Remain," dated March 24, 2017, available at fes.bcaresearch.com. 4 Please see BCA U.S. Equity Strategy Weekly Report, "Too Good To Be True?" dated January 22, 2018, available at uses.bcaresearch.com. Current Recommendations Current Trades Size And Style Views Favor value over growth. Stay neutral small over large caps (downgrade alert).

Special Report

Dear Client, In addition to this Special Report written by my colleagues Mark McClellan and Brian Piccioni, we are sending you an abbreviated weekly report. Best regards, Peter Berezin, Chief Global Strategist Global Investment Strategy Highlights Media reports warn of a "Robot Apocalypse" that is already laying waste to jobs and depressing wages on a broad scale. Technological advance in the past has not prevented improving living standards or led to ever rising joblessness over the decades, but pessimists argue that recent advances are different. The issue is important for financial markets. If structural factors such as automation are holding back inflation by more than in previous decades, then the Fed will have to proceed very slowly in raising rates. We see no compelling evidence that the displacement effect of emerging technologies is any stronger than in the past. Robot usage has had a modest positive impact on overall productivity. Despite this contribution, overall productivity growth has been dismal over the past decade. If automation is increasing 'exponentially' and displacing workers on a broad scale as some claim, one would expect to see accelerating productivity growth, robust capital spending and more violent shifts in occupational shares. Exactly the opposite has occurred. Periods of strong growth in automation have historically been associated with robust, not lackluster, wage gains, contrary to the consensus view. The Fed was successful in meeting the 2% inflation target on average from 2000 to 2007, when the impact of the IT revolution on productivity (and costs) was stronger than that of robot automation today. This and other evidence suggest that it is difficult to make the case that robots will make it tougher for central banks to reach their inflation goals than did previous technological breakthroughs. For investors, this means that we cannot rely on automation to keep inflation depressed irrespective of how tight labor markets become. Feature Recent breakthroughs in technology are awe-inspiring and unsettling. These advances are viewed with great trepidation by many because of the potential to replace humans in the production process. Hype over robots is particularly shrill. Media reports warn of a "Robot Apocalypse" that is already laying waste to jobs and depressing wages on a broad scale. In the first in our series of Special Reports focusing on the structural factors that might be preventing central banks from reaching their inflation targets, we demonstrated that the impact of Amazon is overstated in the press. We estimated that E-commerce is depressing inflation in the U.S. by a mere 0.1 to 0.2 percentage points. This Special Report tackles the impact of automation. We are optimistic that robot technology and artificial intelligence will significantly boost future productivity, and thus reduce costs. But, is there any evidence at the macro level that robot usage has been more deflationary than technological breakthroughs in the past and is, thus, a major driver of the low inflation rates we observe today across the major countries? The question matters, especially for the outlook for central bank policy and the bond market. If structural factors are indeed holding back inflation by more than in previous decades, then the Fed will have to proceed very slowly in raising rates. However, if low inflation simply reflects long lags between wages and the tightening labor market, then inflation may suddenly lurch to life as it has at the end of past cycles. The bond market is not priced for that scenario. Are Robots Different? A Special Report from BCA's Technology Sector Strategy service suggested that the "robot revolution" could be as transformative as previous General Purpose Technologies (GPT), including the steam engine, electricity and the microchip.1 GPTs are technologies that radically alter the economy's production process and make a major contribution to living standards over time. The term "robot" can have different meanings. The most basic definition is "a device that automatically performs complicated and often repetitive tasks," and this encompasses a broad range of machines: From the Jacquard Loom, which was invented over 200 years ago, on to Numerically Controlled (NC) mills and lathes, pick and place machines used in the manufacture of electronics, Autonomous Vehicles (AVs), and even homicidal robots from the future such as the Terminator. Our Technology Sector report made the case that there is nothing particularly sinister about robots. They are just another chapter in a long history of automation. Nor is the displacement of workers unprecedented. The industrial revolution was about replacing human craft labor with capital (machines), which did high-volume work with better quality and productivity. This freed humans for work which had not yet been automated, along with designing, producing and maintaining the machinery. Agriculture offers a good example. This sector involved over 50% of the U.S. labor force until the late 1800s. Steam and then internal combustion-powered tractors, which can be viewed as "robotic horses," contributed to a massive rise in output-per-man hour. The number of hours worked to produce a bushel of wheat fell by almost 98% from the mid-1800s to 1955. This put a lot of farm hands out of work, but these laborers were absorbed over time in other growing areas of the economy. It is the same story for all other historical technological breakthroughs. Change is stressful for those directly affected, but rising productivity ultimately lifts average living standards. Robots will be no different. As we discuss below, however, the increasing use of robots and AI may have a deeper and longer-lasting impact on inequality. Strong Tailwinds Chart 1Robots Are Getting Cheaper Factory robots have improved immensely due to cheaper and more capable control and vision systems. As these systems evolve, the abilities of robots to move around their environment while avoiding obstacles will improve, as will their ability to perform increasingly complex tasks. Most importantly, robots are already able to do more than just routine tasks, thus enabling them to replace or aid humans in higher-skilled processes. Robot prices are also falling fast, especially after quality-adjusting the data (Chart 1). Units are becoming easier to install, program and operate. These trends will help to reduce the barriers-to-entry for the large, untapped, market of small and medium sized enterprises. Robots also offer the ability to do low-volume "customized" production and still keep unit costs low. In the future, self-learning robots will be able to optimize their own performance by analyzing the production of other robots around the world. Robot usage is growing quickly according to data collected by the International Federation of Robotics (IFR) that covers 23 countries. Industrial robot sales worldwide increased to almost 300,000 units in 2016, up 16% from the year before (Chart 2). The stock of industrial robots globally has grown at an annual average pace of 10% since 2010, reaching slightly more than 1.8 million units in 2016.2 Robot usage is far from evenly distributed across industries. The automotive industry is the major consumer of industrial robots, holding 45% of the total stock in 2016 (Chart 3). The computer & electronics industry is a distant second at 17%. Metals, chemicals and electrical/electronic appliances comprise the bulk of the remaining stock. Chart 2Global Robot Usage Chart 3Global Robot Usage By Industry (2016) As far as countries go, Japan has traditionally been the largest market for robots in the world. However, sales have been in a long-term downtrend and the stock of robots has recently been surpassed by China, which has ramped up robot purchases in recent years (Chart 4). Robot density, which is the stock of robots per 10 thousand employed in manufacturing, makes it easier to compare robot usage across countries (Chart 5, panel 2). By this measure, China is not a heavy user of robots compared to other countries. South Korea stands at the top, well above the second-place finishers (Germany and Japan). Large automobile sectors in these three countries explain their high relative robot densities. Chart 4Stock Of Robots By Country (I) Chart 5Stock Of Robots By Country (II) (2016) While the growth rate of robot usage is impressive, it is from a very low base (outside of the automotive industry). The average number of robots per 10,000 employees is only 74 for the 23 countries in the IFR database. Robot use is tiny compared to total man hours worked. In the U.S., spending on robots is only about 5% of total business spending on equipment and software (Chart 6). To put this into perspective, U.S. spending on information, communication and technology (ICT) equipment represented 35-40% of total capital equipment spending during the tech boom in the 1990s and early 2000s.3 Chart 6U.S. Investment In Robots The bottom line is that there is a lot of hype in the press, but robots are not yet widely used across countries or industries. It will be many years before business spending on robots approaches the scale of the 1990s/2000s IT boom. A Deflationary Impact? As noted above, we view robotics as another chapter in a long history of technological advancements. Pessimists suggest that the latest advances are different because they are inherently more threatening to the overall job market and wage share of total income. If the pessimists are right, what are the theoretical channels though which this would have a greater disinflationary effect relative to previous GPT technologies? Faster Productivity Gains: Enhanced productivity drives down unit labor costs, which may be passed along to other industries (as cheaper inputs) and to the end consumer. More Human Displacement: The jobs created in other areas may be insufficient to replace the jobs displaced by robots, leading to lower aggregate income and spending. The loss of income for labor will simply go to the owners of capital, but the point is that the labor share of income might decline. Deflationary pressures could build as aggregate demand falls short of supply. Even in industries that are slow to automate, just the threat of being replaced by robots may curtail wage demands. Inequality: Some have argued that rising inequality is partly because the spoils of new technologies over the past 20 years have largely gone to the owners of capital. This shift may have undermined aggregate demand because upper income households tend to have a high saving rate, thereby depressing overall aggregate demand and inflationary pressures. The human displacement effect, described above, would exacerbate the inequality effect by transferring income from labor to the owners of capital. 1. Productivity It is difficult to see the benefits of robots on productivity at the economy-wide level. Productivity growth has been abysmal across the major developed countries since the Great Recession, but the productivity slowdown was evident long before Lehman collapsed (Chart 7). The productivity slowdown continued even as automation using robots accelerated after 2010. Chart 7Productivity Collapsed Despite Automation Some analysts argue that lackluster productivity is simply a statistical mirage because of the difficulties in measuring output in today's economy. We will not get into the details of the mismeasurement debate here. We encourage interested clients to read a Special Report by the BCA Global Investment Strategy service entitled "Weak Productivity Growth: Don't Blame The Statisticians." 4 Our colleague Peter Berezin makes the case that the unmeasured utility accruing from free internet services is large, but so was the unmeasured utility from antibiotics, radio, indoor plumbing and air conditioning. He argues that the real reason that productivity growth has slowed is that educational attainment has decelerated and businesses have plucked many of the low-hanging fruit made possible by the IT revolution. Cyclical factors stemming from the Great Recession and financial crisis are also to blame, as capital spending has been slow to recover in most of the advanced economies. Some other factors that help to explain the decline in aggregate productivity are provided in Appendix 1. Nonetheless, the poor aggregate productivity performance does not mean that there are no benefits to using robots. The benefits are evident at the industrial level, where measurement issues are presumably less vexing for statisticians (i.e., it is easier to measure the output of the auto industry, for example, than for the economy as a whole). Chart 8 plots the level of robot density in 2016 with average annual productivity growth since 2004 for 10 U.S. manufacturing industries (robot density is presented in deciles). A loose positive relationship is apparent. Chart 8U.S.: Productivity Vs. Robot Density Academic studies estimate that robots have contributed importantly to economy-wide productivity growth. The Centre for Economic and Business Research (CEBR) estimated that labor productivity growth rises by 0.07 to 0.08 percentage points for every 1% rise in the rate of robot density.5 This implies that robots accounted for roughly 10% of the productivity growth experienced since the early 1990s in the major economies. Another study of 14 industries across 17 countries by the Centre for Economic Performance (CEP) found that robots boosted annual productivity growth by 0.36 percentage points over the 1993-2007 period.6 This is impressive because, if this estimate holds true for the U.S., robots' contribution to the 2½% average annual U.S. total productivity growth over the period was 14%. To put the importance of robotics into historical context, its contribution to productivity so far is roughly on par with that of the steam engine (Chart 9). It falls well short of the 0.6 percentage point annual productivity contribution from the IT revolution. The implication is that, while the overall productivity performance has been dismal since 2007, it would have been even worse in the absence of robots. What does this mean for inflation? According to the "cost push" model of the inflation process, an increase in productivity of 0.36% that is not accompanied by associated wage gains would reduce unit labor costs (ULC) by the same amount. This should trim inflation if the cost savings are passed on to the end consumer, although by less than 0.36% because robots can only depress variable costs, not fixed costs. There indeed appears to be a slight negative relationship between robot density and unit labor costs at the industrial level in the U.S., although the relationship is loose at best (Chart 10). Chart 9GPT Contribution To Productivity Chart 10U.S.: Unit Labor Costs Vs. Robot Density In theory, divergences in productivity across industries should only generate shifts in relative prices, and "cost push" inflation dynamics should only operate in the short term. Most economists believe that inflation is a purely monetary phenomenon in the long run, which means that central banks should be able to offset positive productivity shocks by lowering interest rates enough that aggregate demand keeps up with supply. Indeed, the Fed was successful in meeting the 2% inflation target on average from 2000 to 2007, when the impact of the IT revolution on productivity (and costs) was stronger than that of robot automation today. Also, note that inflation is currently low across the major advanced economies, irrespective of the level of robot intensity (Chart 11). From this perspective, it is hard to see that robots should take much of the credit for today's low inflation backdrop. Chart 11Inflation Vs. Robot Density 2. Human Displacement A key question is whether robots and humans are perfect substitutes. If new technologies introduced in the past were perfect substitutes, then it would have led to massive underemployment and all of the income in the economy would eventually have migrated to the owners of capital. The fact that average real household incomes have risen over time, and that there has been no secular upward trend in unemployment rates over the centuries, means that new technologies were at least partly complementary with labor (i.e., the jobs lost as a direct result of productivity gains were more than replaced in other areas of the economy over time). Rather than replacing workers, in many cases tech made humans more productive in their jobs. Rising productivity lifted income and thereby led to the creation of new jobs in other areas. The capital that workers bring to the production process - the skills, know-how and special talents - became more valuable as interaction with technology increased. Like today, there were concerns in the 1950s and 1960s that computerization would displace many types of jobs and lead to widespread idleness and falling household income. With hindsight, there was little to worry about. Some argue that this time is different. Futurists frequently assert that the pace of innovation is not just accelerating, it is accelerating 'exponentially'. Robots can now, or will soon be able to, replace humans in tasks that require cognitive skills. This means that they will be far less complementary to humans than in the past. The displacement effect could thus be much larger, especially given the impressive advances in artificial intelligence. However, Box 1 discusses why the threat to workers posed by AI is also heavily overblown in the media. The CEP multi-country study cited above did not find a large displacement effect; robot usage did not affect the overall number of hours worked in the 23 countries studied (although it found distributional effects - see below). In other words, rather than suppressing overall labor input, robot usage has led to more output, higher productivity, more jobs and stronger wage and income growth. A report by the Economic Policy Institute (EPI)7 takes a broader look at automation, using productivity growth and capital spending as proxies. Automation is what occurs as the implementation of new technologies is incorporated along with new capital equipment or software to replace human labor in the workplace. If automation is increasing 'exponentially' and displacing workers on a broad scale, one would expect to see accelerating productivity growth, robust capital spending, and more violent shifts in occupational shares. Exactly the opposite has occurred. Indeed, the report demonstrates that occupational employment shifts were far slower in the 2000-2015 period than in any decade in the 1900s (Chart 12). Box 1 The Threat From AI Is Overblown Media coverage of AI/Deep Learning has established a consensus view that we believe is well off the mark. A recent Special Report from BCA's Technology Sector Strategy service dispels the myths surrounding AI.8 We believe the consensus, in conjunction with warnings from a variety of sources, is leading to predictions, policy discussions, and even career choices based on a flawed premise. It is worth noting that the most vocal proponents of AI as a threat to jobs and even humanity are not AI experts. At the root of this consensus is the false view that emerging AI technology is anything like true intelligence. Modern AI is not remotely comparable in function to a biological brain. Scientists have a limited understanding of how brains work, and it is unlikely that a poorly understood system can be modeled on a computer. The misconception of intelligence is amplified by headlines claiming an AI "taught itself" a particular task. No AI has ever "taught itself" anything: All AI results have come about after careful programming by often PhD-level experts, who then supplied the system with vast amounts of high quality data to train it. Often these systems have been iterated a number of times and we only hear of successes, not the failures. The need for careful preparation of the AI system and the requirement for high quality data limits the applicability of AI to specific classes of problems where the application justifies the investment in development and where sufficient high-quality data exists. There may be numerous such applications but doubtless many more where AI would not be suitable. Similarly, an AI system is highly adapted to a single problem, or type of problem, and becomes less useful when its application set is expanded. In other words, unlike a human whose abilities improve as they learn more things, an AI's performance on a particular task declines as it does more things. There is a popular misconception that increased computing power will somehow lead to ever improving AI. It is the algorithm which determines the outcome, not the computer performance: Increased computing power leads to faster results, not different results. Advanced computers might lead to more advanced algorithms, but it is pointless to speculate where that may lead: A spreadsheet from 2001 may work faster today but it still gives the same answer. In any event, it is worth noting that a tool ceases to be a tool when it starts having an opinion: there is little reason to develop a machine capable of cognition even if that were possible. Chart 12U.S. Job Rotation Has Slowed The EPI report also notes that these indicators of automation increased rapidly in the late 1990s and early 2000s, a period that saw solid wage growth for American workers. These indicators weakened in the two periods of stagnant wage growth: from 1973 to 1995 and from 2002 to the present. Thus, there is no historical correlation between increases in automation and wage stagnation. Rather than automation, the report argues that it was China's entry into the global trading system that was largely responsible for the hollowing out of the U.S. manufacturing sector. We have also made this argument in previous research. The fact that the major advanced economies are all at, or close to, full employment supports the view that automation has not been an overwhelming headwind for job creation. Chart 13 demonstrates that there has been no relationship between the change in robot density and the loss of manufacturing jobs since 1993. Japan is an interesting case study because it is on the leading edge of the problems associated with an aging population. Interestingly, despite a worsening labor shortage, robot density among Japanese firms is falling. Moreover, the Japanese data show that the industries that have a high robot usage tend to be more, not less, generous with wages than the robot laggard industries. Please see Appendix 2 for more details. Chart 13Global Manufacturing Jobs Vs. Robot Density The bottom line is that it does not appear that labor displacement related to automation has been responsible in any meaningful way for the lackluster average real income growth in the advanced economies since 2007. 3. Inequality That said, there is evidence suggesting that robots are having important distributional effects. The CEP study found that robot use has reduced hours for low-skilled and (to a lesser extent) middle-skilled workers relative to the highly skilled. This finding makes sense conceptually. Technological change can exacerbate inequality by either increasing the relative demand for skilled over unskilled workers (so-called "skill-biased" technological change), or by inducing companies to substitute machinery and other forms of physical capital for workers (so-called "capital-biased" technological change). The former affects the distribution of labor income, while the latter affects the share of income in GDP that labor receives. A Special Report appearing in this publication in 2014 focused on the relationship between technology and inequality.9 The report highlighted that much of the recent technological change has been skill-biased, which heavily favors workers with the talent and education to perform cognitively-demanding tasks, even as it reduces demand for workers with only rudimentary skills. Moreover, technological innovations and globalization increasingly allow the most talented individuals to market their skills to a much larger audience, thus bidding up their wages. The evidence suggests that faster productivity growth leads to higher average real wages and improved living standards, at least over reasonably long horizons. Nonetheless, technological change can, and in the future almost certainly will, increase income inequality. The poor will gain, but not as much as the rich. The fact that higher-income households tend to maintain a higher savings rate than low-income households means that the shift in the distribution of income toward the higher-income households will continue to modestly weigh on aggregate demand. Can the distribution effect be large enough to have a meaningful depressing impact on inflation? We believe that it has played some role in the lackluster recovery since the Great Recession, with the result that an extended period of underemployment has delivered a persistent deflationary impulse in the major developed economies. However, as discussed above, stimulative monetary policy has managed to overcome the impact of inequality and other headwinds on aggregate demand, and has returned the major countries roughly to full employment. Indeed, this year will be the first since 2007 that the G20 economies as a group will be operating slightly above a full employment level. Inflation should respond to excess demand conditions, irrespective of any ongoing demand headwind stemming from inequality. Conclusions Technological change has led to rising living standards over the decades. It did not lead to widespread joblessness and did not prevent central banks from meeting their inflation targets over time. The pessimists argue that this time is different because robots/AI have a much larger displacement effect. Perhaps it will be 20 years before we will know the answer. But our main point is that we have found no evidence that recent advances in robotics and AI, while very impressive, will be any different in their macro impact. There is little evidence that the modern economy is less capable in replacing the jobs lost to automation, although the nature of new technologies may be affecting the distribution of income more than in the past. Real incomes for the middle- and lower-income classes have been stagnant for some time, but this is partly due to productivity growth that is too low, not too high. Moreover, it is not at all clear that positive productivity shocks are disinflationary beyond the near term. The link between robot usage and unit labor costs over the past couple of decades is loose at best at the industry level, and is non-existent when looking across the major countries. The Fed was able to roughly meet its 2% inflation target in the 1990s and the first half of the 2000s, despite IT's impressive contribution to productivity growth during that period. For investors, this means that we cannot rely on automation to keep inflation depressed irrespective of how tight labor markets become. The global output gap will shift into positive territory this year for the first time since the Great Recession. Any resulting rise in inflation will come as a shock since the bond market has discounted continued low inflation for as far as the eye can see. We expect bond yields and implied volatility to rise this year, which may undermine risk assets in the second half. Mark McClellan Senior Vice President The Bank Credit Analyst Brian Piccioni Vice President Technology Sector Strategy Appendix 1 Why Is Productivity So Low? A recent study by the OECD10 reveals that, while frontier firms are charging ahead, there is a widening gap between these firms and the laggards. The study analyzed firm-level data on labor productivity and total factor productivity for 24 countries. "Frontier" firms are defined to be those with productivity in the top 5%. These firms are 3-4 times as productive as the remaining 95%. The authors argue that the underlying cause of this yawning gap is that the diffusion rate of new technologies from the frontier firms to the laggards has slowed within industries. This could be due to rising barriers to entry, which has reduced contestability in markets. Curtailing the creative-destruction process means that there is less pressure to innovate. Barriers to entry may have increased because "...the importance of tacit knowledge as a source of competitive advantage for frontier firms may have risen if increasingly complex technologies were to increase the amount and sophistication of complementary investments required for technological adoption." 11 The bottom line is that aggregate productivity is low because the robust productivity gains for the tech-savvy frontier companies are offset by the long tail of firms that have been slow to adopt the latest technology. Indeed, business spending has been especially weak in this expansion. Chart 14 highlights that the slowdown in U.S. productivity growth has mirrored that of the capital stock. Chart 14U.S. Capex Shortfall Partly To Blame For Poor Productivity Appendix 2 Japan - The Leading Edge Japan is an interesting case study because it is on the leading edge of the problems associated with an aging population. The popular press is full of stories of how robots are taking over. If the stories are to be believed, robots are the answer to the country's shrinking workforce. Robots now serve as helpers for the elderly, priests for weddings and funerals, concierges for hotels and even sexual partners (don't ask). Prime Minister Abe's government has launched a 5-year push to deepen the use of intelligent machines in manufacturing, supply chains, construction and health care. Indeed, Japan was the leader in robotics use for decades. Nonetheless, despite all the hype, Japan's stock of industrial robots has actually been eroding since the late 1990s (Chart 4). Numerous surveys show that firms plan to use robots more in the future because of the difficulty in hiring humans. And there is huge potential: 90% of Japanese firms are small- and medium-sized (SME) and most are not currently using robots. Yet, there has been no wave of robot purchases as of 2016. One problem is the cost; most sophisticated robots are simply too expensive for SMEs to consider. This suggests that one cannot blame robots for Japan's lack of wage growth. The labor shortage has become so acute that there are examples of companies that have turned down sales due to insufficient manpower. Possible reasons why these companies do not offer higher wages to entice workers are beyond the scope of this report. But the fact that the stock of robots has been in decline since the late 1990s does not support the view that Japanese firms are using automation on a broad scale to avoid handing out pay hikes. Indeed, Chart 15 highlights that wage deflation has been the greatest in industries that use almost no robots. Highly automated industries, such as Transportation Equipment and Electronics, have been among the most generous. This supports the view that the productivity afforded by increased robot usage encourages firms to pay their workers more. Looking ahead, it seems implausible that robots can replace all the retiring Japanese workers in the years to come. The workforce will shrink at an annual average pace of 0.33% between 2020 and 2030, according to the Japan Institute for Labour Policy and Training. Productivity growth would have to rise by the same amount to fully offset the dwindling number of workers. But that would require a surge in robot density of 4.1, assuming that each rise in robot density of one adds 0.08% to the level of productivity (Chart 16). The level of robot sales would have to jump by a whopping 2½ times in the first year and continue to rise at the same pace each year thereafter to make this happen. Of course, the productivity afforded by new robots may accelerate in the coming years, but the point is that robot usage would likely have to rise astronomically to offset the impact of the shrinking population. Chart 15Japan: Earnings Vs. Robot Density Chart 16Japan: Where Is The Flood Of Robots? The implication is that, as long as the Japanese economy continues to grow above roughly 1%, the labor market will continue to tighten and wage rates will eventually begin to rise. 1 Please see Technology Sector Strategy Special Report "The Coming Robotics Revolution," dated May 16, 2017, available at tech.bcaresearch.com 2 Note that this includes only robots used in manufacturing industry, and thus excludes robots used in the service sector and households. However, robot usage in services is quite limited and those used in households do not add to GDP. 3 Note that ICT investment and capital stock data includes robots. 4 Please see BCA Global Investment Strategy Special Report "Weak Productivity Growth: Don't Blame The Statisticians," dated March 25, 2016, available at gis.bcaresearch.com 5 Centre for Economic and Business Research (January 2017) "The Impact of Automation." A Report for Redwood. In this report, robot density is defined to be the number of robots per million hours worked. 6 Graetz, G., and Michaels, G. (2015): "Robots At Work." CEP Discussion Paper No 1335. 7 Mishel, L., and Bivens, J. (2017): "The Zombie Robot Argument Lurches On," Economic Policy Institute. 8 Please see BCA Technology Sector Strategy Special Report "Bad Information - Why Misreporting Deep Learning Advances Is A Problem," dated January 9, 2018, available at tech.bcaresearch.com 9 Please see The Bank Credit Analyst, "Rage Against the Machines: Is Technology Exacerbating Inequality?" dated June 2014, available at bca.bcaresearch.com. 10 OECD Productivity Working Papers, No. 05 (2016) "The Best Versus the Rest: The Global Productivity Slowdown, Divergence Across Firms and the Role of Public Policy." 11 Please refer to page 8.

Underweight United Airlines spooked the market this week when they announced their plans to grow capacity by 4-6% per year until 2020; the stock fell by 11% that day and took the S&P airlines index down with it. Capacity additions of this magnitude force competitors to choose between matching or ceding market share. Either choice bodes poorly for airfare pricing, probably unwisely considering how consumer spending on airfares has already fallen off a cliff (second panel). Meanwhile, input prices have shot upward and have diverged sharply from airlines’ ability to pass through fuel costs (third panel). This means a reversal of the downward trend in margins remains well beyond the horizon (bottom panel). Investors should avoid the turbulence; stay underweight. The ticker symbols for the stocks in this index are: BLBG: S5AIRL - DAL, LUV, AAL, UAL, ALK.

Bank Credit Analyst

Highlights U.S. equities 'melted up' in January as tax cuts made the robust growth/low inflation sweet spot even sweeter. Ominously, recent market action is beginning to resemble a classic late cycle blow-off phase. The fundamentals supporting the market will persist through most of the year, before an economic downturn in the U.S. takes hold in 2019. The repatriation of overseas corporate cash will also flatter EPS growth this year via buyback and M&A activity. The S&P 500 could return 14% or more this year. Unfortunately, the consensus now shares our upbeat view for 2018. Valuation is stretched and many indicators suggest that investors have become downright giddy. This month we compare valuation across the major asset classes. U.S. equities are the most overvalued, followed by gold, raw industrials and EM assets. Oil is still close to fair value. Long-term investors should already be scaling back on risk assets. Investors with a 6-12 month horizon should stay overweight equities versus bonds for now, but a risk management approach means that they should not try to squeeze out the last few percentage points of return. In terms of the sequencing of the exit from risk, the most consistent lead/lag relationship relative to previous tops in the equity market is provided by U.S. corporate bonds. For this reason, we are likely to take profits on corporates before equities. EM assets are already at underweight. We still see a window for the U.S. dollar to appreciate, although by only about 5%. A lot of good news is discounted in the euro, peripheral core inflation is slowing and ECB policymakers are getting nervous. Monetary policy remains the main risk to a pro-cyclical investment stance, although not because of the coming change in the makeup of the FOMC. The economy and inflation should justify four Fed rate hikes in 2018 no matter the makeup. The bond bear phase will continue. Feature Chart I-1Investors Are Giddy U.S. equities 'melted up' in January as tax cuts made the robust growth/low inflation sweet spot even sweeter. Ominously, though, recent market action is beginning to resemble the classic late cycle blow-off phase. Such blow-offs can be highly profitable, but also make it more difficult to properly time the market top. Our base case is that the fundamentals supporting the market will persist through most of the year, before an economic downturn in the U.S. takes hold in 2019. Unfortunately, the consensus now shares our upbeat view for 2018 and many indicators suggest that investors have become downright giddy (Chart I-1). These indicators include investor sentiment, our speculation index, and the bull-to-bear ratio. Net S&P earnings revisions and the U.S. economic surprise index are also extremely elevated, while equity and bond implied volatility are near all-time lows. From a contrarian perspective, these observations suggest that a lot of good news is discounted and that the market is vulnerable to even slight disappointments. It is also a bad sign that our Revealed Preference Indicator moved off of its bullish equity signal in January (see Section III for more details). Meanwhile, central banks are beginning to take away the punchbowl as global economic slack dissipates. This is all late-cycle stuff. Equity valuation does not help investors time the peak in markets, but it does tell us something about downside risk and medium-term expected returns. The Shiller P/E ratio has surged above 30 (Chart I-2). Chart I-3 highlights that, historically, average total returns were negligible over the subsequent 10-year period when the Shiller P/E was in the 30-40 range. Granted, the Shiller P/E will likely fall mechanically later this year as the collapse of earnings in 2008 begins to drop out of the 10-year EPS calculation. Nonetheless, even the BCA Composite Valuation indicator, which includes some metrics that account for extremely low bond yields, surpassed +1 standard deviations in January (our threshold for overvaluation; Chart I-2, bottom panel). An overvaluation signal means that investors should be biased to take profits early. Chart I-2BCA Valuation Indicator Surpasses One Sigma Chart I-3Expected Returns Given Starting Point Shiller P/E As we highlighted in our 2018 Outlook Report, long-term investors should already be scaling back on risk assets. We recommend that investors with a 6-12 month horizon should stay overweight equities versus bonds for now, but we need to be vigilant in terms of scouring for signals to take profits. A risk management approach means that investors should not try to get the last few percentage points of return before the peak. U.S. Earnings And Repatriation Before we turn to the timing and sequence of our exit from risk assets, we will first update our thoughts on the earnings cycle. Fourth quarter U.S. earnings season is still in its early innings, but the banking sector has set an upbeat tone. S&P 500 profits are slated to register a 12% growth rate for both Q4/2017 and calendar 2017. Current year EPS growth estimates have been aggressively ratcheted higher (from 12% growth to 16%) in a mere three weeks on the back of Congress' cut to the corporate tax rate.1 U.S. margins fell slightly in the fourth quarter, but remain at a high level on the back of decent corporate pricing power. A pick-up in productivity growth into year-end helped as well. Our short-term profit model remains extremely upbeat (Chart I-4). The positive profit outlook for the first half of the year is broadly based across sectors as well, according to the recently updated EPS forecast models from BCA's U.S. Equity Sector Strategy service.2 The repatriation of overseas corporate cash will also flatter EPS growth this year via buyback and M&A activity. Studies of the 2004 repatriation legislation show that most of the funds "brought home" were paid out to shareholders, mostly in the form of buybacks. A NBER report estimated that for every dollar repatriated, 92 cents was subsequently paid out to shareholders in one form or another. The surge in buybacks occurred in 2005, according to the U.S. Flow of Funds accounts and a proxy using EPS growth less total dollar earnings growth for the S&P 500 (Chart I-5). The contribution to EPS growth from buybacks rose to more than 3 percentage points at the peak in 2005. Chart I-4Profit Growth Still Accelerating Chart I-5U.S. Buybacks To Lift EPS We expect that most of the repatriated funds will again flow through to shareholders, rather than be used to pay down debt or spent on capital goods. Cash has not been a constraint to capital spending in recent years outside of perhaps the small business sector, which has much less to gain from the tax holiday. A revival in animal spirits and capital spending is underway, but this has more to do with the overall tax package and global growth than the ability of U.S. companies to repatriate overseas earnings. Estimates of how much the repatriation could boost EPS vary widely. Most of it will occur in the Tech and Health Care sectors. Buybacks appear to have lifted EPS growth by roughly one percentage point over the past year. We would not be surprised to see this accelerate by 1-2 percentage points, although the timing could be delayed by a year if the 2004 tax holiday provides the correct timeline. This is certainly positive for the equity market, but much of the impact could already be discounted in prices. Organic earnings growth, and the economic and policy outlook will be the main drivers of equity market returns over the next year. We expect some profit margin contraction later this year, but our 5% EPS growth forecast is beginning to look too conservative. This is especially the case because it does not include the corporate tax cuts. The amount by which the tax cuts will boost earnings on an after-tax basis is difficult to estimate, but we are using 5% as a conservative estimate. Adding 2% for buybacks and 2% for dividends, the S&P 500 could provide an attractive 14% total return this year (assuming no multiple expansion). Timing The Exit Chart I-6Timing The Exit (I) That said, we noted in last month's Report and in BCA's 2018 Outlook that this will be a transition year. We expect a recession in the U.S. sometime in 2019 as the Fed lifts rates into restrictive territory. Equities and other risk assets will sniff out the recession about six months in advance, which means that investors should be preparing to take profits sometime during the next 12 months. Last month we discussed some of the indicators we will watch to help us time the exit. The 2/10 Treasury yield curve has been a reliable recession indicator in the past. However, the lead time on the peak in stocks was quite extended at times (Chart I-6). A shift in the 10-year TIPS breakeven rate above 2.4% would be consistent with the Fed's 2% target for the PCE measure of inflation. This would be a signal that the FOMC will have to step-up the pace of rate hikes and aggressively slow economic growth. We expect the Fed to tighten four times in 2018. We are likely to take some money off the table if core inflation is rising, even if it is still below 2%, at the time that the TIPS breakeven reaches 2.4%. We will also be watching seven indicators that we have found to be useful in heralding market tops, which are summarized in our Scorecard Indicator (Chart I-7). At the moment, four out of the seven indicators are positive (Chart I-8): State of the Business Cycle: As early signals that the economy is softening, watch for the ISM new orders minus inventories indicator to slip below zero, or the 3-month growth rate of unemployment claims to rise above zero. Monetary and Financial Conditions: Using interest rates to judge the stance of monetary policy has been complicated by central banks' use of their balance sheet as a policy tool. Thus, it is better to use two of our proprietary indicators: the BCA Monetary Indicator (MI) and the Financial Conditions Indictor. The S&P 500 index has historically rallied strongly when the MI is above its long-term average. Similarly, equities tend to perform well when the FCI is above its 250-day moving average. The MI is sending a negative signal because interest rates have increased and credit growth has slowed. However, the broader FCI remains well in 'bullish' territory. Price Momentum: We simply use the S&P 500 relative to its 200-day moving average to measure momentum. Currently, the index is well above that level, providing a bullish signal for the Scorecard. Sentiment: Our research shows that stock returns have tended to be highest following periods when sentiment is bearish but improving. In contrast, returns have tended to be lowest following periods when sentiment is bullish but deteriorating. The Scorecard includes the BCA Speculation Indicator to capture sentiment, but virtually all measures of sentiment are very high. The next major move has to be down by definition. Thus, sentiment is assigned a negative value in the Scorecard. Value: As discussed above, value is poor based on the Shiller P/E and the BCA Composite Valuation indicator. Valuation may not help with timing, but we include it in our Scorecard because an overvalued signal means investors should err on the side of getting out early. Chart I-7Equity ScoreCard: Watch For A Dip Below 3 Chart I-8Timing The Exit (II) We demonstrated in previous research that a Scorecard reading of three or above was historically associated with positive equity total returns in subsequent months. A drop below three this year would signal the time to de-risk. Table I-1Exit Checklist To our Checklist we add the U.S. Leading Economic index, which has a good track record of calling recessions. However, we will use the LEI excluding the equity market, since we are using it as an indicator for the stock market. It is bullish at the moment. Our Global LEI is also flashing green. Table I-1 provides a summary checklist for trimming equity exposure. At the moment, 2 out of 9 indicators are bearish. Cross Asset Valuation Comparison Clients have asked our view on the appropriate order in which to scale out of risk assets. One way to approach the question is to compare valuation across asset classes. Presumably, the ones that are most overvalued are at greatest risk, and thus profits should be taken the earliest. It is difficult to compare valuation across asset classes. Should one use fitted values from models or simple deviations from moving averages? Over what time period? Since there is no widely accepted approach, we include multiple measures. More than one time period was used in some cases to capture regime changes. Table I-2 provides out 'best guestimate' for nine asset classes. The approaches range from sophisticated methods developed over many years (i.e. our equity valuation indicators), to regression analysis on the fundamentals (oil), to simple deviations from a time trend (real raw industrial commodity prices and gold). Table I-2Valuation Levels For Major Asset Classes We averaged the valuation readings in cases where there are multiple estimates for a single asset class. The results are shown in Chart I-9. Chart I-9Valuation Levels For Major Asset Classes U.S. equities stand out as the most expensive by far, at 1.8 standard deviations above fair value. Gold, raw industrials and EM equities are next at one standard deviation overvalued. EM sovereign bond spreads come next at 0.7, followed closely by U.S. Treasurys (real yield levels) and investment-grade corporate (IG) bonds (expressed as a spread). High-yield (HY) is only about 0.3 sigma expensive, based on default-adjusted spreads over the Treasury curve. That said, both IG and HY are quite expensive in absolute terms based on the fact that government bonds are expensive. Oil is sitting very close to fair value, despite the rapid price run up over the past couple of months. This makes oil exposure doubly attractive at the moment because the fundamentals point to higher prices at a time when the underlying asset is not expensive. Sequencing Around Past S&P 500 Peaks Historical analysis around equity market peaks provides an alternative approach to the sequencing question. Table I-3 presents the number of days that various asset classes peaked before or after the past major five tops in the S&P 500. A negative number indicates that the asset class peaked before U.S. equities, and a positive number means that it peaked after. Table I-3Asset Class Leads & Lags Vs. Peak In S&P 500 Unfortunately, there is no consistent pattern observed for EM equities, raw industrials, U.S. cyclical stocks, Tech stocks, or small-cap versus large-cap relative returns. Sometimes they peaked before the S&P 500, and sometime after. The EM sovereign bond excess return index peaked about 130 days in advance of the 1998 and 2007 U.S. equity market tops, although we only have three episodes to analyse due to data limitations. Oil is a mixed bag. A peak in the price of gold led the equity market in four out of five episodes, but the lead time is long and variable. The most consistent lead/lag relationship is given by the U.S. corporate bond market. Both investment- and speculative-grade excess returns relative to government bonds peaked in advance of U.S. stocks in four of the five episodes. High-yield excess returns provided the most lead time, peaking on average 154 days in advance. Excess returns to high-yield were a better signal than total returns. This leading relationship is one reason why we plan to trim exposure to corporate bonds within our bond portfolio in advance of scaling back on equities. But the 'return of vol' that we expect to occur later this year will take a toll on carry trades more generally. We are already underweight EM equities and bonds. This EM recommendation has not gone in our favor, but it would make little sense to upgrade them now given our positive views on volatility and the dollar. An unwinding of carry trades will also hit the high-yielding currencies outside of the EM space, such as the Kiwi and Aussie dollar. Base metal prices will be hit particularly hard if the 2019 U.S. recession spills over to the EM economies as we expect. We may downgrade base metals from neutral to underweight around the time that we downgrade equities, but much depends on the evolution of the Chinese economy in the coming months. Oil is a different story. OPEC 2.0 is likely to cut back on supply in the face of an economic downturn, helping to keep prices elevated. We therefore may not trim energy exposure this year. As for equity sectors, our recommended portfolio is still overweight cyclicals for now. Our synchronized global capex boom, rising bond yield, and firm oil price themes keep us overweight the Industrials, Energy and Financial sectors. Utilities and Homebuilders are underweight. Tech is part of the cyclical sector, but poor valuation keeps us underweight. That said, our sector specialists are already beginning a gradual shift away from cyclicals toward defensives for risk management purposes. This transition will continue in the coming months as we de-risk. We are also shifting small caps to neutral on earnings disappointments and elevated debt levels. The Dollar Pain Trade Market shifts since our last publication have largely gone in our favor; stocks have surged, corporate bonds spreads have tightened, oil prices have spiked, bonds have sold off and cyclical stocks have outperformed defensives. One area that has gone against us is the U.S. dollar. Relative interest rate expectations have moved in favor of the dollar as we expected at both the short- and long-ends of the curve. Nonetheless, the dollar has not tracked its historical relationship versus both the yen and euro. The Greenback did not even get a short-term boost from the passage of the tax plan and holiday on overseas earnings. Perhaps this is because the lion's share of "overseas" earnings are already held in U.S. dollars. Reportedly, a large fraction is even held in U.S. banks on U.S. territory. Currency conversion is thus not a major bullish factor for the U.S. dollar. The recent bout of dollar weakness began around the time of the release of the ECB Minutes in January which were interpreted as hawkish because they appeared to be preparing markets for changes in monetary policy. The European debt crisis and economic recession were the reasons for the ECB's asset purchases and negative interest rate policy. Neither of these conditions are in place now. The ECB is meeting as we go to press, and we expect some small adjustments in the Statement that remove references to the need for "crisis" level accommodations. Subsequent steps will be to prepare markets for a complete end to QE, perhaps in September, and then for rates hikes likely in 2019. The key point is that European monetary policy has moved beyond 'peak stimulus' and the normalization process will continue. Perhaps this is partly to blame for euro strength although, as mentioned above, interest rate differentials have moved in favor of the dollar. Does this mean that the dollar has peaked and has entered a cyclical bear phase that will persist over the next 6-12 months? The answer is 'no', although we are less bullish than in the past. We believe there is still a window for the dollar to appreciate against the euro and in broader trade-weighted terms by about 5%. First, a lot of euro-bullish news has been discounted (Chart I-10). Positive economic surprises heavily outstripped that in the U.S. last year, but that phase is now over. The euro appears expensive based on interest rate differentials, and euro sentiment is close to a bullish extreme. This all suggests that market positioning has become a negative factor for the currency. Chart I-10Euro: A Lot Of Bullish News Is Discounted Second, the chorus of complaints against the euro's strength is growing among European central bankers, including Ewald Nowotny, the rather hawkish Austrian central banker. Policymakers' concerns may partly reflect the fact that peripheral inflation excluding food and energy has already weakened to 0.6% from a high of 1.3% in April last year (Chart I-10, fourth panel). Third, U.S. consumer price and wage inflation have yet to pick up meaningfully. The dollar should receive a lift if core U.S. inflation clearly moves toward the Fed's 2% target, as we expect. The FOMC would suddenly appear to have fallen behind the curve and U.S. rate expectations would ratchet higher. Chart I-10, bottom panel, highlights that the euro will weaken if U.S. core inflation rises versus that in the Eurozone. The implication is that the Euro's appreciation has progressed too far and is due for a pullback. As for the yen, the currency surged in January when the Bank of Japan (BoJ) announced a reduction in long-dated JGB purchases. This simply acknowledged what has already occurred. It was always going to be impossible to target both the quantity of bond purchases and the level of 10-year yield simultaneously. Keeping yields near the target required less purchases than they thought. The market interpreted the BoJ's move as a possible prelude to lifting the 10-year yield target. It is perhaps not surprising that the market took the news this way. The economy is performing extremely well; our model that incorporates high-frequency economic data suggests that real GDP growth will move above 3% in the coming quarters. The Japanese economy is benefiting from the end of a fiscal drag and from a rebound in EM growth. Nonetheless, following January's BoJ policy meeting, Kuroda poured cold water on speculation that the BoJ may soon end or adjust the YCC. Recent speeches by BoJ officials reinforce the view that the MPC wants to see an overshoot of actual inflation that will lower real interest rates and thereby reinforce the strong economic activity that is driving higher inflation. Only then will officials be convinced that their job is done. Given that inflation excluding food and energy only stands at 0.3%, the BoJ is still a long way from the overshoot it desires. On the positive side, Japan's large current account surplus and yen undervaluation provide underlying support for the currency. Balancing the offsetting positive and negative forces, our foreign exchange strategists have shifted to neutral on the yen. The Euro remains underweight while the dollar is overweight. Similar to our dollar view, we still see a window for U.S. Treasurys to underperform the global hedged fixed-income benchmark as world bond yields shift higher this year. European government bonds will also sell off, but should outperform Treasurys. JGBs will provide the best refuge for bondholders during the global bond bear phase, since the BoJ will prevent a rise in yields inside of the 10-year maturity. Our global bond strategists upgraded U.K. gilts to overweight in January. Momentum in the U.K. economy is slowing, as a weaker consumer, slower housing activity, and softer capital spending are offsetting a pickup in exports. With the inflationary impulse from the 2016 plunge in the Pound now fading, and with Brexit uncertainty weighing on business confidence, the Bank of England will struggle to raise rates in 2018. FOMC Transition Monetary policy remains the main risk to a pro-cyclical investment stance, although not because of the coming change in the makeup of the FOMC. An abrupt shift in policy is unlikely. There was some support at the December 2017 FOMC meeting to study the use of nominal GDP or price level targeting as a policy framework, but this has been an ongoing debate that will likely continue for years to come. The Fed will remain committed to its current monetary policy framework once Powell takes over. Table I-4 provides a summary of who will be on the FOMC next year, including their policy bias. Chart I-11 compares the recent FOMC makeup with the coming Powell FOMC (voting members only). The hawk/dove ratio will not change much under Powell, unless Trump stacks the vacant spots with hawks. Table I-4Composition Of The FOMC Chart I-11Composition Of Voting FOMC Members 2017 Vs. 2018 In any event, history shows that the FOMC strives to avoid major shifts in policy around changeovers in the Fed Chair. In previous transitions, the previous path for rates was maintained by an average of 13 months. Moreover, Powell has shown that he is not one to rock the boat during his time on the FOMC. It will be the evolution of the economy and inflation, not the composition of the FOMC, that will have the biggest impact on markets at the end of the day. Recent speeches reveal that policymakers across the hawk/dove spectrum are moving modesty toward the hawkish side because growth has accelerated at a time when unemployment is already considered to be below full-employment by many policymakers. The melt-up in equity indexes in January did little to calm worries about financial excesses either. The Fed is struggling to understand the strength of the structural factors that could be holding down inflation. This month's Special Report, beginning on page 21, focusses on the impact of robot automation. While advances on this front are impressive, we conclude that it is difficult to find evidence that robots are more deflationary than previous technological breakthroughs. Thus, increased robot usage should not prevent inflation from rising as the labor market continues to tighten. The macro backdrop will likely justify the FOMC hiking at least as fast as the dots currently forecast. The risks are skewed to the upside. The median Fed dot calls for an unemployment rate of 3.9% by end-2018, only marginally lower than today's rate of 4.1%. This is inconsistent with real GDP growth well in excess of its supply-side potential. The unemployment rate is more likely to reach a 49-year low of 3.5% by the end of this year. As highlighted in last month's Report, a key risk to the bull market in risk assets is the end of the 'low vol/low rate' world. The selloff in the bond market in January may mark the start of this process. Conclusions We covered a lot of ground in this month's Overview of the markets, so we will keep the conclusions brief and focused on the risks. Our key point is that the fundamentals remain positive for risk assets, but that a lot of good news is discounted and it appears that we have entered a classic blow-off phase. This will be a transition year to a recession in the U.S. in 2019. Given that valuation for most risk assets is quite stretched, and given that the monetary taps are starting to close, investors must plan for the exit and keep an eye on our timing checklist. The main risk to our pro-cyclical portfolio is a rise in U.S. inflation and the Fed's response, which we believe will end the sweet spot for risk assets. Apart from this, our geopolitical strategists point to several other items that could upset the applecart this year:3 1. Trade China has cooperated with the U.S. in trying to tame North Korea. Nonetheless, President Trump is committed to an "America First" trade policy and he may need to show some muscle against China ahead of the midterm elections in November in order to rally his base. It is politically embarrassing to the Administration that China racked up its largest trade surplus ever with the U.S. in Trump's first year in office. A key question is whether the President goes after China via a series of administrative rulings - such as the recently announced tariffs on solar panels and white goods - or whether he applies an across-the-board tariff and/or fine. The latter would have larger negative macroeconomic implications. 2. Iran On January 12, President Trump threatened not to waive sanctions against Iran the next time they come due (May 12), unless some new demands are met. Pressure from the U.S. President comes at a delicate time for Iran. Domestic unrest has been ongoing since December 28. Although protests have largely fizzled out, they have reopened the rift between the clerical regime, led by Supreme Leader Ayatollah Ali Khamenei, and moderate President Hassan Rouhani. Iranian hardliners, who control part of the armed forces, could lash out in the Persian Gulf, either by threatening to close the Straits of Hormuz or by boarding foreign vessels in international waters. The domestic political calculus in both Iran and the U.S. make further Tehran-Washington tensions likely. For the time being, however, we expect only a minor geopolitical risk premium to seep into the energy markets, supporting our bullish House View on oil prices. 3. China Last month's Special Report highlighted that significant structural reforms are on the way in China, now that President Xi has amassed significant political support for his reform agenda. The reforms should be growth-positive in the long term, but could be a net negative for growth in the near term depending on how deftly the authorities handle the monetary and fiscal policy dials. The risk is that the authorities make a policy mistake by staying too tight, as occurred in 2015. We are monitoring a number of indicators that should warn if a policy mistake is unfolding. On this front, January brought some worrying economic data. The latest figures for both nominal imports and money growth slowed. Given that M2 and M3 are components of BCA's Li Keqiang Leading Indicator, and that nominal imports directly impact China's contribution to global growth, this raises the question of whether December's economic data suggest that China is slowing at a more aggressive pace than we expect. For now, our answer is no. First, China's trade numbers are highly volatile; nominal import growth remains elevated after smoothing the data. Second, China's export growth remains buoyant, consistent with a solid December PMI reading. The bottom line is that we are sticking with our view that China will experience a benign deceleration in terms of its impact on DM risk assets, but we will continue to monitor the situation closely. Mark McClellan Senior Vice President The Bank Credit Analyst January 25, 2018 Next Report: February 22, 2018 1 According to Thomson Reuters/IBES. 2 Please see U.S. Equity Sector Strategy Special Report "White Paper: Introducing Our U.S. Equity Sector Earnings Models," dated January 16, 2018, available at uses.bcaresearch.com 3 For more information, please see BCA Geopolitical Strategy Weekly Report "Upside Risks In U.S., Downside Risks In China," dated January 17, 2018, available at gps.bcaresearch.com. Also see "Watching Five Risks," dated January 24, 2018. II. The Impact Of Robots On Inflation Media reports warn of a "Robot Apocalypse" that is already laying waste to jobs and depressing wages on a broad scale. Technological advance in the past has not prevented improving living standards or led to ever rising joblessness over the decades, but pessimists argue that recent advances are different. The issue is important for financial markets. If structural factors such as automation are holding back inflation by more than in previous decades, then the Fed will have to proceed very slowly in raising rates. We see no compelling evidence that the displacement effect of emerging technologies is any stronger than in the past. Robot usage has had a modest positive impact on overall productivity. Despite this contribution, overall productivity growth has been dismal over the past decade. If automation is increasing 'exponentially' and displacing workers on a broad scale as some claim, one would expect to see accelerating productivity growth, robust capital spending and more violent shifts in occupational shares. Exactly the opposite has occurred. Periods of strong growth in automation have historically been associated with robust, not lackluster, wage gains, contrary to the consensus view. The Fed was successful in meeting the 2% inflation target on average from 2000 to 2007, when the impact of the IT revolution on productivity (and costs) was stronger than that of robot automation today. This and other evidence suggest that it is difficult to make the case that robots will make it tougher for central banks to reach their inflation goals than did previous technological breakthroughs. For investors, this means that we cannot rely on automation to keep inflation depressed irrespective of how tight labor markets become. Recent breakthroughs in technology are awe-inspiring and unsettling. These advances are viewed with great trepidation by many because of the potential to replace humans in the production process. Hype over robots is particularly shrill. Media reports warn of a "Robot Apocalypse" that is already laying waste to jobs and depressing wages on a broad scale. In the first in our series of Special Reports focusing on the structural factors that might be preventing central banks from reaching their inflation targets, we demonstrated that the impact of Amazon is overstated in the press. We estimated that E-commerce is depressing inflation in the U.S. by a mere 0.1 to 0.2 percentage points. This Special Report tackles the impact of automation. We are optimistic that robot technology and artificial intelligence will significantly boost future productivity, and thus reduce costs. But, is there any evidence at the macro level that robot usage has been more deflationary than technological breakthroughs in the past and is, thus, a major driver of the low inflation rates we observe today across the major countries? The question matters, especially for the outlook for central bank policy and the bond market. If structural factors are indeed holding back inflation by more than in previous decades, then the Fed will have to proceed very slowly in raising rates. However, if low inflation simply reflects long lags between wages and the tightening labor market, then inflation may suddenly lurch to life as it has at the end of past cycles. The bond market is not priced for that scenario. Are Robots Different? A Special Report from BCA's Technology Sector Strategy service suggested that the "robot revolution" could be as transformative as previous General Purpose Technologies (GPT), including the steam engine, electricity and the microchip.1 GPTs are technologies that radically alter the economy's production process and make a major contribution to living standards over time. The term "robot" can have different meanings. The most basic definition is "a device that automatically performs complicated and often repetitive tasks," and this encompasses a broad range of machines: From the Jacquard Loom, which was invented over 200 years ago, on to Numerically Controlled (NC) mills and lathes, pick and place machines used in the manufacture of electronics, Autonomous Vehicles (AVs), and even homicidal robots from the future such as the Terminator. Our Technology Sector report made the case that there is nothing particularly sinister about robots. They are just another chapter in a long history of automation. Nor is the displacement of workers unprecedented. The industrial revolution was about replacing human craft labor with capital (machines), which did high-volume work with better quality and productivity. This freed humans for work which had not yet been automated, along with designing, producing and maintaining the machinery. Agriculture offers a good example. This sector involved over 50% of the U.S. labor force until the late 1800s. Steam and then internal combustion-powered tractors, which can be viewed as "robotic horses," contributed to a massive rise in output-per-man hour. The number of hours worked to produce a bushel of wheat fell by almost 98% from the mid-1800s to 1955. This put a lot of farm hands out of work, but these laborers were absorbed over time in other growing areas of the economy. It is the same story for all other historical technological breakthroughs. Change is stressful for those directly affected, but rising productivity ultimately lifts average living standards. Robots will be no different. As we discuss below, however, the increasing use of robots and AI may have a deeper and longer-lasting impact on inequality. Strong Tailwinds Chart II-1Robots Are Getting Cheaper Factory robots have improved immensely due to cheaper and more capable control and vision systems. As these systems evolve, the abilities of robots to move around their environment while avoiding obstacles will improve, as will their ability to perform increasingly complex tasks. Most importantly, robots are already able to do more than just routine tasks, thus enabling them to replace or aid humans in higher-skilled processes. Robot prices are also falling fast, especially after quality-adjusting the data (Chart II-1). Units are becoming easier to install, program and operate. These trends will help to reduce the barriers-to-entry for the large, untapped, market of small and medium sized enterprises. Robots also offer the ability to do low-volume "customized" production and still keep unit costs low. In the future, self-learning robots will be able to optimize their own performance by analyzing the production of other robots around the world. Robot usage is growing quickly according to data collected by the International Federation of Robotics (IFR) that covers 23 countries. Industrial robot sales worldwide increased to almost 300,000 units in 2016, up 16% from the year before (Chart II-2). The stock of industrial robots globally has grown at an annual average pace of 10% since 2010, reaching slightly more than 1.8 million units in 2016.2 Robot usage is far from evenly distributed across industries. The automotive industry is the major consumer of industrial robots, holding 45% of the total stock in 2016 (Chart II-3). The computer & electronics industry is a distant second at 17%. Metals, chemicals and electrical/electronic appliances comprise the bulk of the remaining stock. Chart II-2Global Robot Usage Chart II-3Global Robot Usage By Industry (2016) As far as countries go, Japan has traditionally been the largest market for robots in the world. However, sales have been in a long-term downtrend and the stock of robots has recently been surpassed by China, which has ramped up robot purchases in recent years (Chart II-4). Robot density, which is the stock of robots per 10 thousand employed in manufacturing, makes it easier to compare robot usage across countries (Chart II-5, panel 2). By this measure, China is not a heavy user of robots compared to other countries. South Korea stands at the top, well above the second-place finishers (Germany and Japan). Large automobile sectors in these three countries explain their high relative robot densities. Chart II-4Stock Of Robots By Country (I) Chart II-5Stock Of Robots By Country (II) (2016) While the growth rate of robot usage is impressive, it is from a very low base (outside of the automotive industry). The average number of robots per 10,000 employees is only 74 for the 23 countries in the IFR database. Robot use is tiny compared to total man hours worked. Chart II-6U.S. Investment In Robots In the U.S., spending on robots is only about 5% of total business spending on equipment and software (Chart II-6). To put this into perspective, U.S. spending on information, communication and technology (ICT) equipment represented 35-40% of total capital equipment spending during the tech boom in the 1990s and early 2000s.3 The bottom line is that there is a lot of hype in the press, but robots are not yet widely used across countries or industries. It will be many years before business spending on robots approaches the scale of the 1990s/2000s IT boom. A Deflationary Impact? As noted above, we view robotics as another chapter in a long history of technological advancements. Pessimists suggest that the latest advances are different because they are inherently more threatening to the overall job market and wage share of total income. If the pessimists are right, what are the theoretical channels though which this would have a greater disinflationary effect relative to previous GPT technologies? Faster Productivity Gains: Enhanced productivity drives down unit labor costs, which may be passed along to other industries (as cheaper inputs) and to the end consumer. More Human Displacement: The jobs created in other areas may be insufficient to replace the jobs displaced by robots, leading to lower aggregate income and spending. The loss of income for labor will simply go to the owners of capital, but the point is that the labor share of income might decline. Deflationary pressures could build as aggregate demand falls short of supply. Even in industries that are slow to automate, just the threat of being replaced by robots may curtail wage demands. Inequality: Some have argued that rising inequality is partly because the spoils of new technologies over the past 20 years have largely gone to the owners of capital. This shift may have undermined aggregate demand because upper income households tend to have a high saving rate, thereby depressing overall aggregate demand and inflationary pressures. The human displacement effect, described above, would exacerbate the inequality effect by transferring income from labor to the owners of capital. 1. Productivity It is difficult to see the benefits of robots on productivity at the economy-wide level. Productivity growth has been abysmal across the major developed countries since the Great Recession, but the productivity slowdown was evident long before Lehman collapsed (Chart II-7). The productivity slowdown continued even as automation using robots accelerated after 2010. Chart II-7Productivity Collapsed Despite Automation Some analysts argue that lackluster productivity is simply a statistical mirage because of the difficulties in measuring output in today's economy. We will not get into the details of the mismeasurement debate here. We encourage interested clients to read a Special Report by the BCA Global Investment Strategy service entitled "Weak Productivity Growth: Don't Blame The Statisticians." 4 Our colleague Peter Berezin makes the case that the unmeasured utility accruing from free internet services is large, but so was the unmeasured utility from antibiotics, radio, indoor plumbing and air conditioning. He argues that the real reason that productivity growth has slowed is that educational attainment has decelerated and businesses have plucked many of the low-hanging fruit made possible by the IT revolution. Cyclical factors stemming from the Great Recession and financial crisis are also to blame, as capital spending has been slow to recover in most of the advanced economies. Some other factors that help to explain the decline in aggregate productivity are provided in Appendix II-1. Nonetheless, the poor aggregate productivity performance does not mean that there are no benefits to using robots. The benefits are evident at the industrial level, where measurement issues are presumably less vexing for statisticians (i.e., it is easier to measure the output of the auto industry, for example, than for the economy as a whole). Chart II-8 plots the level of robot density in 2016 with average annual productivity growth since 2004 for 10 U.S. manufacturing industries (robot density is presented in deciles). A loose positive relationship is apparent. Chart II-8U.S.: Productivity Vs. Robot Density Academic studies estimate that robots have contributed importantly to economy-wide productivity growth. The Centre for Economic and Business Research (CEBR) estimated that labor productivity growth rises by 0.07 to 0.08 percentage points for every 1% rise in the rate of robot density.5 This implies that robots accounted for roughly 10% of the productivity growth experienced since the early 1990s in the major economies. Another study of 14 industries across 17 countries by the Centre for Economic Performance (CEP) found that robots boosted annual productivity growth by 0.36 percentage points over the 1993-2007 period.6 This is impressive because, if this estimate holds true for the U.S., robots' contribution to the 2½% average annual U.S. total productivity growth over the period was 14%. To put the importance of robotics into historical context, its contribution to productivity so far is roughly on par with that of the steam engine (Chart II-9). It falls well short of the 0.6 percentage point annual productivity contribution from the IT revolution. The implication is that, while the overall productivity performance has been dismal since 2007, it would have been even worse in the absence of robots. What does this mean for inflation? According to the "cost push" model of the inflation process, an increase in productivity of 0.36% that is not accompanied by associated wage gains would reduce unit labor costs (ULC) by the same amount. This should trim inflation if the cost savings are passed on to the end consumer, although by less than 0.36% because robots can only depress variable costs, not fixed costs. There indeed appears to be a slight negative relationship between robot density and unit labor costs at the industrial level in the U.S., although the relationship is loose at best (Chart II-10). Chart II-9GPT Contribution To Productivity Chart II-10U.S.: Unit Labor Costs Vs. Robot Density In theory, divergences in productivity across industries should only generate shifts in relative prices, and "cost push" inflation dynamics should only operate in the short term. Most economists believe that inflation is a purely monetary phenomenon in the long run, which means that central banks should be able to offset positive productivity shocks by lowering interest rates enough that aggregate demand keeps up with supply. Indeed, the Fed was successful in meeting the 2% inflation target on average from 2000 to 2007, when the impact of the IT revolution on productivity (and costs) was stronger than that of robot automation today. Also, note that inflation is currently low across the major advanced economies, irrespective of the level of robot intensity (Chart II-11). From this perspective, it is hard to see that robots should take much of the credit for today's low inflation backdrop. Chart II-11Inflation Vs. Robot Density 2. Human Displacement A key question is whether robots and humans are perfect substitutes. If new technologies introduced in the past were perfect substitutes, then it would have led to massive underemployment and all of the income in the economy would eventually have migrated to the owners of capital. The fact that average real household incomes have risen over time, and that there has been no secular upward trend in unemployment rates over the centuries, means that new technologies were at least partly complementary with labor (i.e., the jobs lost as a direct result of productivity gains were more than replaced in other areas of the economy over time). Rather than replacing workers, in many cases tech made humans more productive in their jobs. Rising productivity lifted income and thereby led to the creation of new jobs in other areas. The capital that workers bring to the production process - the skills, know-how and special talents - became more valuable as interaction with technology increased. Like today, there were concerns in the 1950s and 1960s that computerization would displace many types of jobs and lead to widespread idleness and falling household income. With hindsight, there was little to worry about. Some argue that this time is different. Futurists frequently assert that the pace of innovation is not just accelerating, it is accelerating 'exponentially'. Robots can now, or will soon be able to, replace humans in tasks that require cognitive skills. This means that they will be far less complementary to humans than in the past. The displacement effect could thus be much larger, especially given the impressive advances in artificial intelligence. However, Box II-1 discusses why the threat to workers posed by AI is also heavily overblown in the media. The CEP multi-country study cited above did not find a large displacement effect; robot usage did not affect the overall number of hours worked in the 23 countries studied (although it found distributional effects - see below). In other words, rather than suppressing overall labor input, robot usage has led to more output, higher productivity, more jobs and stronger wage and income growth. A report by the Economic Policy Institute (EPI)7 takes a broader look at automation, using productivity growth and capital spending as proxies. Automation is what occurs as the implementation of new technologies is incorporated along with new capital equipment or software to replace human labor in the workplace. If automation is increasing 'exponentially' and displacing workers on a broad scale, one would expect to see accelerating productivity growth, robust capital spending, and more violent shifts in occupational shares. Exactly the opposite has occurred. Indeed, the report demonstrates that occupational employment shifts were far slower in the 2000-2015 period than in any decade in the 1900s (Chart II-12). Box II-1 The Threat From AI Is Overblown Media coverage of AI/Deep Learning has established a consensus view that we believe is well off the mark. A recent Special Report from BCA's Technology Sector Strategy service dispels the myths surrounding AI.8 We believe the consensus, in conjunction with warnings from a variety of sources, is leading to predictions, policy discussions, and even career choices based on a flawed premise. It is worth noting that the most vocal proponents of AI as a threat to jobs and even humanity are not AI experts. At the root of this consensus is the false view that emerging AI technology is anything like true intelligence. Modern AI is not remotely comparable in function to a biological brain. Scientists have a limited understanding of how brains work, and it is unlikely that a poorly understood system can be modeled on a computer. The misconception of intelligence is amplified by headlines claiming an AI "taught itself" a particular task. No AI has ever "taught itself" anything: All AI results have come about after careful programming by often PhD-level experts, who then supplied the system with vast amounts of high quality data to train it. Often these systems have been iterated a number of times and we only hear of successes, not the failures. The need for careful preparation of the AI system and the requirement for high quality data limits the applicability of AI to specific classes of problems where the application justifies the investment in development and where sufficient high-quality data exists. There may be numerous such applications but doubtless many more where AI would not be suitable. Similarly, an AI system is highly adapted to a single problem, or type of problem, and becomes less useful when its application set is expanded. In other words, unlike a human whose abilities improve as they learn more things, an AI's performance on a particular task declines as it does more things. There is a popular misconception that increased computing power will somehow lead to ever improving AI. It is the algorithm which determines the outcome, not the computer performance: Increased computing power leads to faster results, not different results. Advanced computers might lead to more advanced algorithms, but it is pointless to speculate where that may lead: A spreadsheet from 2001 may work faster today but it still gives the same answer. In any event, it is worth noting that a tool ceases to be a tool when it starts having an opinion: there is little reason to develop a machine capable of cognition even if that were possible. Chart II-12U.S. Job Rotation Has Slowed The EPI report also notes that these indicators of automation increased rapidly in the late 1990s and early 2000s, a period that saw solid wage growth for American workers. These indicators weakened in the two periods of stagnant wage growth: from 1973 to 1995 and from 2002 to the present. Thus, there is no historical correlation between increases in automation and wage stagnation. Rather than automation, the report argues that it was China's entry into the global trading system that was largely responsible for the hollowing out of the U.S. manufacturing sector. We have also made this argument in previous research. The fact that the major advanced economies are all at, or close to, full employment supports the view that automation has not been an overwhelming headwind for job creation. Chart II-13 demonstrates that there has been no relationship between the change in robot density and the loss of manufacturing jobs since 1993. Japan is an interesting case study because it is on the leading edge of the problems associated with an aging population. Interestingly, despite a worsening labor shortage, robot density among Japanese firms is falling. Moreover, the Japanese data show that the industries that have a high robot usage tend to be more, not less, generous with wages than the robot laggard industries. Please see Appendix II-2 for more details. Chart II-13Global Manufacturing Jobs Vs. Robot Density The bottom line is that it does not appear that labor displacement related to automation has been responsible in any meaningful way for the lackluster average real income growth in the advanced economies since 2007. 3. Inequality That said, there is evidence suggesting that robots are having important distributional effects. The CEP study found that robot use has reduced hours for low-skilled and (to a lesser extent) middle-skilled workers relative to the highly skilled. This finding makes sense conceptually. Technological change can exacerbate inequality by either increasing the relative demand for skilled over unskilled workers (so-called "skill-biased" technological change), or by inducing companies to substitute machinery and other forms of physical capital for workers (so-called "capital-biased" technological change). The former affects the distribution of labor income, while the latter affects the share of income in GDP that labor receives. A Special Report appearing in this publication in 2014 focused on the relationship between technology and inequality.9 The report highlighted that much of the recent technological change has been skill-biased, which heavily favors workers with the talent and education to perform cognitively-demanding tasks, even as it reduces demand for workers with only rudimentary skills. Moreover, technological innovations and globalization increasingly allow the most talented individuals to market their skills to a much larger audience, thus bidding up their wages. The evidence suggests that faster productivity growth leads to higher average real wages and improved living standards, at least over reasonably long horizons. Nonetheless, technological change can, and in the future almost certainly will, increase income inequality. The poor will gain, but not as much as the rich. The fact that higher-income households tend to maintain a higher savings rate than low-income households means that the shift in the distribution of income toward the higher-income households will continue to modestly weigh on aggregate demand. Can the distribution effect be large enough to have a meaningful depressing impact on inflation? We believe that it has played some role in the lackluster recovery since the Great Recession, with the result that an extended period of underemployment has delivered a persistent deflationary impulse in the major developed economies. However, as discussed above, stimulative monetary policy has managed to overcome the impact of inequality and other headwinds on aggregate demand, and has returned the major countries roughly to full employment. Indeed, this year will be the first since 2007 that the G20 economies as a group will be operating slightly above a full employment level. Inflation should respond to excess demand conditions, irrespective of any ongoing demand headwind stemming from inequality. Conclusions Technological change has led to rising living standards over the decades. It did not lead to widespread joblessness and did not prevent central banks from meeting their inflation targets over time. The pessimists argue that this time is different because robots/AI have a much larger displacement effect. Perhaps it will be 20 years before we will know the answer. But our main point is that we have found no evidence that recent advances in robotics and AI, while very impressive, will be any different in their macro impact. There is little evidence that the modern economy is less capable in replacing the jobs lost to automation, although the nature of new technologies may be affecting the distribution of income more than in the past. Real incomes for the middle- and lower-income classes have been stagnant for some time, but this is partly due to productivity growth that is too low, not too high. Moreover, it is not at all clear that positive productivity shocks are disinflationary beyond the near term. The link between robot usage and unit labor costs over the past couple of decades is loose at best at the industry level, and is non-existent when looking across the major countries. The Fed was able to roughly meet its 2% inflation target in the 1990s and the first half of the 2000s, despite IT's impressive contribution to productivity growth during that period. For investors, this means that we cannot rely on automation to keep inflation depressed irrespective of how tight labor markets become. The global output gap will shift into positive territory this year for the first time since the Great Recession. Any resulting rise in inflation will come as a shock since the bond market has discounted continued low inflation for as far as the eye can see. We expect bond yields and implied volatility to rise this year, which may undermine risk assets in the second half. Mark McClellan Senior Vice President The Bank Credit Analyst Brian Piccioni Vice President Technology Sector Strategy Appendix II-1 Why Is Productivity So Low? A recent study by the OECD10 reveals that, while frontier firms are charging ahead, there is a widening gap between these firms and the laggards. The study analyzed firm-level data on labor productivity and total factor productivity for 24 countries. "Frontier" firms are defined to be those with productivity in the top 5%. These firms are 3-4 times as productive as the remaining 95%. The authors argue that the underlying cause of this yawning gap is that the diffusion rate of new technologies from the frontier firms to the laggards has slowed within industries. This could be due to rising barriers to entry, which has reduced contestability in markets. Curtailing the creative-destruction process means that there is less pressure to innovate. Barriers to entry may have increased because "...the importance of tacit knowledge as a source of competitive advantage for frontier firms may have risen if increasingly complex technologies were to increase the amount and sophistication of complementary investments required for technological adoption." 11 The bottom line is that aggregate productivity is low because the robust productivity gains for the tech-savvy frontier companies are offset by the long tail of firms that have been slow to adopt the latest technology. Indeed, business spending has been especially weak in this expansion. Chart II-14 highlights that the slowdown in U.S. productivity growth has mirrored that of the capital stock. Chart II-14U.S. Capex Shortfall Partly To Blame For Poor Productivity Appendix II-2 Japan - The Leading Edge Japan is an interesting case study because it is on the leading edge of the problems associated with an aging population. The popular press is full of stories of how robots are taking over. If the stories are to be believed, robots are the answer to the country's shrinking workforce. Robots now serve as helpers for the elderly, priests for weddings and funerals, concierges for hotels and even sexual partners (don't ask). Prime Minister Abe's government has launched a 5-year push to deepen the use of intelligent machines in manufacturing, supply chains, construction and health care. Indeed, Japan was the leader in robotics use for decades. Nonetheless, despite all the hype, Japan's stock of industrial robots has actually been eroding since the late 1990s (Chart II-4). Numerous surveys show that firms plan to use robots more in the future because of the difficulty in hiring humans. And there is huge potential: 90% of Japanese firms are small- and medium-sized (SME) and most are not currently using robots. Yet, there has been no wave of robot purchases as of 2016. One problem is the cost; most sophisticated robots are simply too expensive for SMEs to consider. This suggests that one cannot blame robots for Japan's lack of wage growth. The labor shortage has become so acute that there are examples of companies that have turned down sales due to insufficient manpower. Possible reasons why these companies do not offer higher wages to entice workers are beyond the scope of this report. But the fact that the stock of robots has been in decline since the late 1990s does not support the view that Japanese firms are using automation on a broad scale to avoid handing out pay hikes. Indeed, Chart II-15 highlights that wage deflation has been the greatest in industries that use almost no robots. Highly automated industries, such as Transportation Equipment and Electronics, have been among the most generous. This supports the view that the productivity afforded by increased robot usage encourages firms to pay their workers more. Looking ahead, it seems implausible that robots can replace all the retiring Japanese workers in the years to come. The workforce will shrink at an annual average pace of 0.33% between 2020 and 2030, according to the Japan Institute for Labour Policy and Training. Productivity growth would have to rise by the same amount to fully offset the dwindling number of workers. But that would require a surge in robot density of 4.1, assuming that each rise in robot density of one adds 0.08% to the level of productivity (Chart II-16). The level of robot sales would have to jump by a whopping 2½ times in the first year and continue to rise at the same pace each year thereafter to make this happen. Of course, the productivity afforded by new robots may accelerate in the coming years, but the point is that robot usage would likely have to rise astronomically to offset the impact of the shrinking population. Chart II-15Japan: Earnings Vs. Robot Density Chart II-16Japan: Where Is The Flood Of Robots? The implication is that, as long as the Japanese economy continues to grow above roughly 1%, the labor market will continue to tighten and wage rates will eventually begin to rise. 1 Please see Technology Sector Strategy Special Report "The Coming Robotics Revolution," dated May 16, 2017, available at tech.bcaresearch.com 2 Note that this includes only robots used in manufacturing industry, and thus excludes robots used in the service sector and households. However, robot usage in services is quite limited and those used in households do not add to GDP. 3 Note that ICT investment and capital stock data includes robots. 4 Please see BCA Global Investment Strategy Special Report "Weak Productivity Growth: Don't Blame The Statisticians," dated March 25, 2016, available at gis.bcaresearch.com 5 Centre for Economic and Business Research (January 2017): "The Impact of Automation." A Report for Redwood. In this report, robot density is defined to be the number of robots per million hours worked. 6 Graetz, G., and Michaels, G. (2015): "Robots At Work." CEP Discussion Paper No 1335. 7 Mishel, L., and Bivens, J. (2017): "The Zombie Robot Argument Lurches On," Economic Policy Institute. 8 Please see BCA Technology Sector Strategy Special Report "Bad Information - Why Misreporting Deep Learning Advances Is A Problem," dated January 9, 2018, available at tech.bcaresearch.com 9 Please see The Bank Credit Analyst, "Rage Against The Machines: Is Technology Exacerbating Inequality?" dated June 2014, available at bca.bcaresearch.com 10 OECD Productivity Working Papers, No. 05 (2016): "The Best Versus the Rest: The Global Productivity Slowdown, Divergence Across Firms and the Role of Public Policy." 11 Please refer to page 27. III. Indicators And Reference Charts As we highlight in the Overview section, the earnings backdrop for the U.S. equity market remains very upbeat, as highlighted by the rise in the net earnings revisions and net earnings surprises indexes. Bottom-up analysts will likely continue to boost after-tax earnings estimates for the year as they adjust to the U.S. tax cut news. Our main concern is that a lot of good news is now discounted. Our Technical Indicator remains bullish, but our composite valuation indicator surpassed one sigma in January, which is our threshold of overvaluation. From these levels of overvaluation, the medium-term outlook for equity total returns is negligible. Our speculation index is at all-time highs and implied volatility is low, underscoring that investors are extremely bullish. From a contrary perspective, this is a warning sign for the equity market. Our Monetary Indicator has also moved further into 'bearish' territory for equities, although overall financial conditions remain positive for growth. It is also disconcerting that our Revealed Preference Indicator (RPI) shifted to a 'sell' signal for stocks, following five straight months on a 'buy' signal. This occurred because investors may be buying based on speculation rather than on a firm belief in the staying power of the underlying fundamentals. For now, though, our Willingness-to-Pay indicator for the U.S. rose sharply in January, highlighting that investor equity inflows are very strong and are favoring U.S. equities relative to Japan and the Eurozone. This is perhaps not surprising given the U.S. tax cuts just passed by Congress. The RPI indicators track flows, and thus provide information on what investors are actually doing, as opposed to sentiment indexes that track how investors are feeling. Our U.S. bond technical indicator shows that Treasurys are close to oversold territory, suggesting that we may be in store for a consolidation period following January's surge in yields. Treasurys are slightly cheap on our valuation metric, although not by enough to justify closing short duration positions. The U.S. dollar is oversold and due for a bounce. EQUITIES: Chart III-1U.S. Equity Indicators Chart III-2Willingness To Pay For Risk Chart III-3U.S. Equity Sentiment Indicators Chart III-4Revealed Preference Indicator Chart III-5U.S. Stock Market Valuation Chart III-6U.S. Earnings Chart III-7Global Stock Market And Earnings: ##br##Relative Performance Chart III-8Global Stock Market And Earnings: ##br##Relative Performance FIXED INCOME: Chart III-9U.S. Treasurys And Valuations Chart III-10U.S. Treasury Indicators Chart III-11Selected U.S. Bond Yields Chart III-1210-Year Treasury Yield ComponentsChart III-13U.S. Corporate Bonds And Health Monitor Chart III-14Global Bonds: Developed Markets Chart III-15Global Bonds: Emerging Markets CURRENCIES: Chart III-16U.S. Dollar And PPP Chart III-17U.S. Dollar And Indicator Chart III-18U.S. Dollar Fundamentals Chart III-19Japanese Yen Technicals Chart III-20Euro Technicals Chart III-21Euro/Yen Technicals Chart III-22Euro/Pound Technicals COMMODITIES: Chart III-23Broad Commodity Indicators Chart III-24Commodity Prices Chart III-25Commodity Prices Chart III-26Commodity Sentiment Chart III-27Speculative Positioning ECONOMY: Chart III-28U.S. And Global Macro Backdrop Chart III-29U.S. Macro Snapshot Chart III-30U.S. Growth Outlook Chart III-31U.S. Cyclical Spending Chart III-32U.S. Labor Market Chart III-33U.S. Consumption Chart III-34U.S. Housing Chart III-35U.S. Debt And Deleveraging Chart III-36U.S. Financial Conditions Chart III-37Global Economic Snapshot: Europe Chart III-38Global Economic Snapshot: China Mark McClellan Senior Vice President The Bank Credit Analyst

Highlights A thorough audit of our trade book highlights that our country and sector allocation recommendations have been quite profitable for investors. Of the 12 active trades in our book, 11 have generated a positive return, including one with a 32% annualized rate of return. A review of the original basis and subsequent performance of our trades suggests that investors should close 6 out of 12 of our active positions, predominantly related to resource & construction and domestic stock market themes. We will be looking for opportunities to add new trades to our book over the coming weeks and months that have broad, "big-picture" relevance. Watch this space. Feature In this week's report we conduct a thorough audit of our trade book, by revisiting the original basis and subsequent performance of all 12 of our active trades. While these trades have been initiated at different points over the past five years, they can be broadly grouped into five different themes: Core Equity Allocation & General Pro-Risk Trades (4 Trades) Reform-Oriented Trades (2 Trades) Resource & Construction Plays (2 Trades) Domestic Stock Market Trades (2 Trades) Trades Linked To Hong Kong (2 Trades) Overall, our trade book performance has been excellent. Of the 12 active trades in our book, 11 have generated a positive return, including one with a 32% annualized rate of return (since December 2015). As a result of our trade book review, we recommend that investors close six trades and maintain six over the coming 6-12 months. The closed trades predominantly fall into the resource & construction and domestic stock market categories, although we also recommend closing our long China H-share / short industrial commodity trade as well as our long Hong Kong REITs / short Hong Kong broad market trade. We present our rationale for retaining or closing each trade below. Over the coming weeks and months we will be looking for opportunities to add new trades to our book. Stay tuned. Core Equity Allocation & General Pro-Risk Trades We have four open core equity allocation and pro-risk trades: Overweight MSCI China Investable stocks versus the emerging markets benchmark, initiated on May 2, 2012 Long China H-shares / short industrial commodities, initiated on March 16, 2016 Short MSCI Taiwan / Long MSCI China Investable, initiated on February 2, 2017 and Long China onshore corporate bonds, initiated on June 22, 2017 We recommend that investors stick with three of these trades, but close the long China H-shares / short industrial commodities position for the following reasons: Chart 1Be Overweight China Vs EM In This Environment Overweight MSCI China Investable Stocks Versus The EM Benchmark (Maintain) This trade represents one of the most important equity allocation calls for Chinese stocks, and is one of the ways that BCA expresses a view on the Chinese economy in our House View Matrix.1 While it hasn't always been the case, we noted in a recent Special Report that Chinese stocks have become a high-beta equity market versus both the global aggregate and the emerging market benchmark, even when excluding the technology sector.2 China's high-beta nature, the fact that EM equities remain in an uptrend (Chart 1), and our view that China's ongoing slowdown is likely to be benign and controlled all suggest that investors should continue to overweight Chinese stocks vs their emerging market peers. Long China H-Shares / Short Industrial Commodities (Close) We initiated this trade in March 2016, one month after Chinese stock prices bottomed following the significant economic slowdown in 2015. At that time it was not clear to global investors that a mini-cycle upswing in the Chinese economy had begun, and this pair trade was a way of taking a limited pro-risk bet. Given our view of a benign, controlled economic slowdown in China, this hedged trade is no longer needed, especially given the uncertain impact of ongoing supply side constraints in China on global commodity prices. As such, we recommend that investors close the trade, locking in an annualized return of 15.7%. Short MSCI Taiwan / Long MSCI China Investable (Maintain) Chart 2If The TWD Declines Materially, ##br##Upgrade Taiwan (From Short) We initiated our short MSCI Taiwan / long MSCI China investable trade last February, when the risk of protectionist action from the Trump administration loomed large. While there have been no negative trade actions levied against Taiwan this year, macro factors, particularly the strength of the currency, continue to argue for an underweight stance within the greater China bourses (China, Hong Kong, and Taiwan). We reviewed the basis of this trade in a report last month,3 and recommended that investors stick with the call despite significantly oversold conditions (Chart 2). A material easing in pressure on Taiwan's trade-weighted exchange rate appears to be the most likely catalyst to close the trade and to upgrade Taiwan within a portfolio of greater China equities. Long China Onshore Corporate Bonds (Maintain) Chinese corporate bond yields have risen materially since late-2016, largely in response to expectations of tighter monetary policy. These expectations have been validated, with 3-month interbank rates having risen over 200bps since late-2016. We argued last summer that the phase of maximum liquidity tightening was likely over, and that quality spreads and government bond yields would probably drop over the coming three to six months. While this clearly did not occur (yields and spreads rose), the total return from this trade has remained in the black owing to the significant yield advantage of these bonds versus similarly-rated bonds in the developed world. Chart 3 highlights that Chinese 5-year corporate bond spreads are also considerably less correlated with equity prices than their investment-grade peers in the U.S. This underscores that the rise in yields and spreads over the past year has reflected expectations of tighter monetary policy, not rising default risk. Our sense is that barring a significant improvement in China's growth momentum, significant further monetary policy tightening is improbable, meaning that corporate bond yields are not likely to rise much further. As a final point, as of today's report we are changing the benchmark for this trade from a BCA calculation based on a basket of 5-year AAA and AA-rated corporate bonds to the ChinaBond Corporate Credit Bond Total Return Index. Chart 3Chinese Corporate Spreads Aren't A Risk ##br##Barometer Like In The U.S. Reform-Oriented Trades We have two open trades related to China's rebooted reform initiative, both of which were initiated on November 16, 2017: Long China investable consumer staples / short consumer discretionary stocks and Long China investable environmental and social governance (ESG) leaders / short investable broad market These trades were recently opened, and we continue to recommend that investors maintain both positions: Long China Investable Consumer Staples / Short Consumer Discretionary Stocks (Maintain) The basis for the first trade stems from the current limitations of China's investable consumer discretionary index as a clear-cut play on retail-oriented consumer spending. We argued in our November 16 Weekly Report that Chinese investable consumer staples would be a better play on Chinese consumer spending owing to the material weight of the automobiles & components industry group in the discretionary sector, which may fare poorly over the coming year due to the environmental mandate of President Xi's proposed reforms. We argued in the report that this trade would likely be driven by alpha rather than beta, and indeed Chart 4 illustrates that staples continue to rise relative to discretionary against a backdrop of a rising broad market. Long China Investable ESG leaders / Short Investable Broad Market (Maintain) In the same report we recommended that investors overweight the China investable ESG leaders index, based on the goal of favoring firms that are best positioned to deliver "sustainable" growth in an era of heightened environmental reforms. The index overweights firms with the highest MSCI ESG ratings in each sector (using a proprietary MSCI ranking scheme), and maintains similar sector weights as the investable benchmark, which limits the beta risk of the trade. Chart 5 highlights that the trade is progressing in line with our expectations, suggesting that investors stick with the position over the coming 6-12 months. Chart 4Staples Vs Discretionary Isn't A Low Beta Trade Chart 5Likely To Continue To Outperform Resource & Construction Plays We have two open trades related to the resource sector: Long China investable oil & gas stocks / short global oil & gas stocks, initiated on April 26, 2014 and Long China investable construction materials sector / short investable broad market, initiated on December 9, 2015 We recommend that investors close both of these positions, based on the following rationale: Chart 6Similar Earnings Profile, ##br##But Weaker Dividend Payouts Long China Investable Oil & Gas Stocks / Short Global Oil & Gas Stocks (Close) This trade was initiated based on the view that the valuation gap between Chinese and global oil & gas companies is unjustifiable given that the earnings off both sectors are globally driven. Indeed, Chart 6 shows that the trailing EPS profiles of both sectors in US$ terms have been broadly similar over the past few years, and yet China's oil & gas sector trades at a 40% price-to-book discount relative to its global peers. However, panel 2 of Chart 6 highlights that this discount may represent investor concerns about earnings quality and/or state-owned corporate governance. The chart shows that while the earnings ROE for Chinese oil & gas companies is higher than that of the global average, the dividend ROE (dividends per share as a percent of shareholders equity) is considerably lower. While China's oil & gas dividend ROE has recently been rising, the gap remains wide relative to global oil & gas companies, suggesting that there is no significant re-rating catalyst that is likely to emerge over the coming 6-12 months. Close for an annualized return of 1.4%. Long China Investable Construction Material Stocks / Short China Investable Broad Market (Close) The relative performance of Chinese investable construction material stocks has been positive over the past two years, with the trade having generated an 8.1% annualized return since initiation. There are two factors contributing to our view that it is time for investors to book profits on this trade. The first is that China's investable construction materials are dominated by cement companies, which may suffer in relative terms from China's rebooted reform initiative this year.4 The second is that the relative performance of construction materials stocks is closely correlated with, and led by, the growth in total real estate investment (Chart 7). Residential investment makes up a significant component of total real estate investment, and Chart 8 highlights that a significant gap between floor space sold and completed has narrowed the inventory to sales ratio over the past three years. But the ratio remains somewhat elevated relative to its history which, when coupled with the ongoing growth slowdown in China and the deceleration in total real estate investment growth, implies a poor risk/reward ratio over the coming 6-12 months. Chart 7Cement Producers Trade Off Of Real Estate Investment Chart 8No Clear Construction Boom Is Imminent Domestic Stock Market Trades We have two open trades related to China's domestic stock market: Long China domestic utility sector / short domestic broad market, initiated on January 22, 2014 and Long China domestic food & beverage sector / short domestic broad market, initiated on December 9, 2015 Similar to our resource & construction plays, we recommend that investors close both of our recommended domestic stock market trades: Long China Domestic Utility Sector / Short Domestic Broad Market (Close) We initiated this trade in early-2014, following a comprehensive reform plan released in late-2013 by the Chinese government. The plan called for allowing market forces to play a decisive role in allocating resources, which we argued would grant utilities more pricing power, reduce their earnings volatility associated with policy risks, and lead to a structural positive re-rating. Chart 9 illustrates that this trade gained significant ground in 2014 and early-2015, even prior to the significant melt-up in domestic stock prices that began in Q2 2015. However, the trade has underperformed significantly since the middle of last year, which has been driven by a sharp deterioration in ROE. This decline in ROE appears to have been cost-driven, as coal is an important feedstock for Chinese utility companies and has risen substantially in price over the past two years. While domestic utilities are now significantly oversold in relative terms, we recommend that investors close this trade because the original reform-oriented basis has shifted significantly. The priorities that emanated from October's Party Congress were decidedly environmental in nature, meaning that coal prices may very well remain elevated over the coming 6-12 months (due to restricted supply). This means that a recovery in ROE would rest on the need to raise utility prices, which is a low-visibility event that will be difficult to predict. Close for an annualized return of 3%. Long China Domestic Food & Beverage Sector / Short Domestic Broad Market (Close) We initiated this trade in December 2015, based on this sector's superior corporate fundamentals and undemanding valuation levels. We argued that the anti-corruption campaign since late-2012 was likely the cause of prior underperformance, given that the group is dominated by a few high-end alcohol producers. The market overacted to the high-profile crackdown, and ultimately the fundamentals of the sector did not deteriorate materially. Our view has panned out spectacularly, with the trade having earned a 32% annualized return since inception5 (Chart 10 panel 1). While the group's ROE remains significantly above that of the domestic benchmark, valuation measures suggest that investors have more than priced this in (Chart 10 panel 2). The trade has mostly played out and we would not like to overstay our welcome. In addition, panel 3 illustrates that technical conditions are extremely overbought, suggesting that investors are being presented with an excellent opportunity to exit the position. Chart 9Sidelined By A Major Hit To ROE Chart 10Time To Book Profits Trades Linked To Hong Kong We have two open trades related to Hong Kong: Long U.S. / short Hong Kong 10-Year government bonds, initiated on January 15, 2014 and Short Hong Kong property investors / long Hong Kong broad market, initiated on January 21, 2015 We recommend that investors stick with the first and close the second, based on the following perspectives: Long U.S. / Short Hong Kong 10-Year Government Bonds (Maintain) Hong Kong has an open capital account and an exchange rate pegged to the U.S. dollar, meaning that its monetary policy is directly tied to that of the U.S. Yet, Hong Kong's 10-year government bond yield is non-trivially below that of the U.S., which argues for a short stance versus similar maturity U.S. Treasurys. While it is true that the Hong Kong - U.S. 10-year yield spread does vary and can widen over a 6-12 month horizon, Chart 11 highlights that the relative total return profile of the trade (in unhedged terms) trends higher over time due to the carry advantage. Short Hong Kong REITs / Long Hong Kong Broad Market (Close) There are cross-currents facing the outlook for Hong Kong REITs vs the broad market, arguing for a neutral rather than an underweight stance. Close this trade for an annualized return of 3.6%. While the relative performance of global REITs is typically negatively correlated with bond yields, Chart 12 shows that the relationship with Hong Kong property yields has been positive and lagging (i.e. falling yields lead declining relative performance, and vice versa). Under this regime, a rise in U.S. government bond yields, as we expect, would suggest an improvement in the relative performance of Hong Kong REITs. Chart 11A Straightforward Carry Pick Up Trade Chart 12Rising Bond Yields Implies ##br##Positive HK REIT Performance Chart 13 highlights that periods of positive yield / REIT performance correlation have tended to occur when Hong Kong property prices are rising significantly relative to income, as they have been for the past several years. One interpretation of this dynamic is that when house prices are overvalued and potentially vulnerable, REIT investors react positively to an improvement in economic fundamentals (which tends to push yields up due to higher interest rate expectations). The risk of an eventual collapse of Hong Kong property prices is clear, but we cannot identify an obvious catalyst for this to occur over the coming 6-12 months. Importantly, the fact that property prices have continued to rise during a period of tighter mainland capital controls suggests that only a significant economic shock will be enough to derail the uptrend in prices, circumstances that we do not expect over the coming year. Finally, Chart 14 highlights that Hong Kong REITs are deeply discounted relative to book value when compared against the broad market. This suggests that at least some of the risks associated with the property market have already been priced in by investors. Chart 13Yields & REITs Positively Correlated ##br##When House Prices Are Overvalued Chart 14Hong Kong REITs Are Cheap Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com Lin Xiang, Research Analyst linx@bcaresearch.com 1 https://www.bcaresearch.com/trades 2 Please see China Investment Strategy Weekly Report, "China: No Longer A Low-Beta Market", dated January 11, 2018, available at cis.bcaresearch.com. 3 Please see China Investment Strategy Weekly Report "Taiwan: Awaiting A Re-Rating Catalyst", dated December 14, 2017, available at cis.bcaresearch.com. 4 Please see China Investment Strategy Weekly Report, "Messages From The Market, Post-Party Congress", dated November 16, 2017, available at cis.bcaresearch.com. 5 Please note that the total return from this trade had been erroneously reported for some time due a data processing error on BCA's part. The return since inception now properly sources the China CSI SWS Food & Beverage index from CHOICE. We sincerely regret the error and any confusion it may have caused. Cyclical Investment Stance Equity Sector Recommendations

Bank Credit Analyst

Media reports warn of a "Robot Apocalypse" that is already laying waste to jobs and depressing wages on a broad scale. Technological advance in the past has not prevented improving living standards or led to ever rising joblessness over the decades, but pessimists argue that recent advances are different. The issue is important for financial markets. If structural factors such as automation are holding back inflation by more than in previous decades, then the Fed will have to proceed very slowly in raising rates. We see no compelling evidence that the displacement effect of emerging technologies is any stronger than in the past. Robot usage has had a modest positive impact on overall productivity. Despite this contribution, overall productivity growth has been dismal over the past decade. If automation is increasing 'exponentially' and displacing workers on a broad scale as some claim, one would expect to see accelerating productivity growth, robust capital spending and more violent shifts in occupational shares. Exactly the opposite has occurred. Periods of strong growth in automation have historically been associated with robust, not lackluster, wage gains, contrary to the consensus view. The Fed was successful in meeting the 2% inflation target on average from 2000 to 2007, when the impact of the IT revolution on productivity (and costs) was stronger than that of robot automation today. This and other evidence suggest that it is difficult to make the case that robots will make it tougher for central banks to reach their inflation goals than did previous technological breakthroughs. For investors, this means that we cannot rely on automation to keep inflation depressed irrespective of how tight labor markets become. Recent breakthroughs in technology are awe-inspiring and unsettling. These advances are viewed with great trepidation by many because of the potential to replace humans in the production process. Hype over robots is particularly shrill. Media reports warn of a "Robot Apocalypse" that is already laying waste to jobs and depressing wages on a broad scale. In the first in our series of Special Reports focusing on the structural factors that might be preventing central banks from reaching their inflation targets, we demonstrated that the impact of Amazon is overstated in the press. We estimated that E-commerce is depressing inflation in the U.S. by a mere 0.1 to 0.2 percentage points. This Special Report tackles the impact of automation. We are optimistic that robot technology and artificial intelligence will significantly boost future productivity, and thus reduce costs. But, is there any evidence at the macro level that robot usage has been more deflationary than technological breakthroughs in the past and is, thus, a major driver of the low inflation rates we observe today across the major countries? The question matters, especially for the outlook for central bank policy and the bond market. If structural factors are indeed holding back inflation by more than in previous decades, then the Fed will have to proceed very slowly in raising rates. However, if low inflation simply reflects long lags between wages and the tightening labor market, then inflation may suddenly lurch to life as it has at the end of past cycles. The bond market is not priced for that scenario. Are Robots Different? A Special Report from BCA's Technology Sector Strategy service suggested that the "robot revolution" could be as transformative as previous General Purpose Technologies (GPT), including the steam engine, electricity and the microchip.1 GPTs are technologies that radically alter the economy's production process and make a major contribution to living standards over time. The term "robot" can have different meanings. The most basic definition is "a device that automatically performs complicated and often repetitive tasks," and this encompasses a broad range of machines: From the Jacquard Loom, which was invented over 200 years ago, on to Numerically Controlled (NC) mills and lathes, pick and place machines used in the manufacture of electronics, Autonomous Vehicles (AVs), and even homicidal robots from the future such as the Terminator. Our Technology Sector report made the case that there is nothing particularly sinister about robots. They are just another chapter in a long history of automation. Nor is the displacement of workers unprecedented. The industrial revolution was about replacing human craft labor with capital (machines), which did high-volume work with better quality and productivity. This freed humans for work which had not yet been automated, along with designing, producing and maintaining the machinery. Agriculture offers a good example. This sector involved over 50% of the U.S. labor force until the late 1800s. Steam and then internal combustion-powered tractors, which can be viewed as "robotic horses," contributed to a massive rise in output-per-man hour. The number of hours worked to produce a bushel of wheat fell by almost 98% from the mid-1800s to 1955. This put a lot of farm hands out of work, but these laborers were absorbed over time in other growing areas of the economy. It is the same story for all other historical technological breakthroughs. Change is stressful for those directly affected, but rising productivity ultimately lifts average living standards. Robots will be no different. As we discuss below, however, the increasing use of robots and AI may have a deeper and longer-lasting impact on inequality. Strong Tailwinds Chart II-1Robots Are Getting Cheaper Factory robots have improved immensely due to cheaper and more capable control and vision systems. As these systems evolve, the abilities of robots to move around their environment while avoiding obstacles will improve, as will their ability to perform increasingly complex tasks. Most importantly, robots are already able to do more than just routine tasks, thus enabling them to replace or aid humans in higher-skilled processes. Robot prices are also falling fast, especially after quality-adjusting the data (Chart II-1). Units are becoming easier to install, program and operate. These trends will help to reduce the barriers-to-entry for the large, untapped, market of small and medium sized enterprises. Robots also offer the ability to do low-volume "customized" production and still keep unit costs low. In the future, self-learning robots will be able to optimize their own performance by analyzing the production of other robots around the world. Robot usage is growing quickly according to data collected by the International Federation of Robotics (IFR) that covers 23 countries. Industrial robot sales worldwide increased to almost 300,000 units in 2016, up 16% from the year before (Chart II-2). The stock of industrial robots globally has grown at an annual average pace of 10% since 2010, reaching slightly more than 1.8 million units in 2016.2 Robot usage is far from evenly distributed across industries. The automotive industry is the major consumer of industrial robots, holding 45% of the total stock in 2016 (Chart II-3). The computer & electronics industry is a distant second at 17%. Metals, chemicals and electrical/electronic appliances comprise the bulk of the remaining stock. Chart II-2Global Robot Usage Chart II-3Global Robot Usage By Industry (2016) As far as countries go, Japan has traditionally been the largest market for robots in the world. However, sales have been in a long-term downtrend and the stock of robots has recently been surpassed by China, which has ramped up robot purchases in recent years (Chart II-4). Robot density, which is the stock of robots per 10 thousand employed in manufacturing, makes it easier to compare robot usage across countries (Chart II-5, panel 2). By this measure, China is not a heavy user of robots compared to other countries. South Korea stands at the top, well above the second-place finishers (Germany and Japan). Large automobile sectors in these three countries explain their high relative robot densities. Chart II-4Stock Of Robots By Country (I) Chart II-5Stock Of Robots By Country (II) (2016) While the growth rate of robot usage is impressive, it is from a very low base (outside of the automotive industry). The average number of robots per 10,000 employees is only 74 for the 23 countries in the IFR database. Robot use is tiny compared to total man hours worked. Chart II-6U.S. Investment In Robots In the U.S., spending on robots is only about 5% of total business spending on equipment and software (Chart II-6). To put this into perspective, U.S. spending on information, communication and technology (ICT) equipment represented 35-40% of total capital equipment spending during the tech boom in the 1990s and early 2000s.3 The bottom line is that there is a lot of hype in the press, but robots are not yet widely used across countries or industries. It will be many years before business spending on robots approaches the scale of the 1990s/2000s IT boom. A Deflationary Impact? As noted above, we view robotics as another chapter in a long history of technological advancements. Pessimists suggest that the latest advances are different because they are inherently more threatening to the overall job market and wage share of total income. If the pessimists are right, what are the theoretical channels though which this would have a greater disinflationary effect relative to previous GPT technologies? Faster Productivity Gains: Enhanced productivity drives down unit labor costs, which may be passed along to other industries (as cheaper inputs) and to the end consumer. More Human Displacement: The jobs created in other areas may be insufficient to replace the jobs displaced by robots, leading to lower aggregate income and spending. The loss of income for labor will simply go to the owners of capital, but the point is that the labor share of income might decline. Deflationary pressures could build as aggregate demand falls short of supply. Even in industries that are slow to automate, just the threat of being replaced by robots may curtail wage demands. Inequality: Some have argued that rising inequality is partly because the spoils of new technologies over the past 20 years have largely gone to the owners of capital. This shift may have undermined aggregate demand because upper income households tend to have a high saving rate, thereby depressing overall aggregate demand and inflationary pressures. The human displacement effect, described above, would exacerbate the inequality effect by transferring income from labor to the owners of capital. 1. Productivity It is difficult to see the benefits of robots on productivity at the economy-wide level. Productivity growth has been abysmal across the major developed countries since the Great Recession, but the productivity slowdown was evident long before Lehman collapsed (Chart II-7). The productivity slowdown continued even as automation using robots accelerated after 2010. Chart II-7Productivity Collapsed Despite Automation Some analysts argue that lackluster productivity is simply a statistical mirage because of the difficulties in measuring output in today's economy. We will not get into the details of the mismeasurement debate here. We encourage interested clients to read a Special Report by the BCA Global Investment Strategy service entitled "Weak Productivity Growth: Don't Blame The Statisticians." 4 Our colleague Peter Berezin makes the case that the unmeasured utility accruing from free internet services is large, but so was the unmeasured utility from antibiotics, radio, indoor plumbing and air conditioning. He argues that the real reason that productivity growth has slowed is that educational attainment has decelerated and businesses have plucked many of the low-hanging fruit made possible by the IT revolution. Cyclical factors stemming from the Great Recession and financial crisis are also to blame, as capital spending has been slow to recover in most of the advanced economies. Some other factors that help to explain the decline in aggregate productivity are provided in Appendix II-1. Nonetheless, the poor aggregate productivity performance does not mean that there are no benefits to using robots. The benefits are evident at the industrial level, where measurement issues are presumably less vexing for statisticians (i.e., it is easier to measure the output of the auto industry, for example, than for the economy as a whole). Chart II-8 plots the level of robot density in 2016 with average annual productivity growth since 2004 for 10 U.S. manufacturing industries (robot density is presented in deciles). A loose positive relationship is apparent. Chart II-8U.S.: Productivity Vs. Robot Density Academic studies estimate that robots have contributed importantly to economy-wide productivity growth. The Centre for Economic and Business Research (CEBR) estimated that labor productivity growth rises by 0.07 to 0.08 percentage points for every 1% rise in the rate of robot density.5 This implies that robots accounted for roughly 10% of the productivity growth experienced since the early 1990s in the major economies. Another study of 14 industries across 17 countries by the Centre for Economic Performance (CEP) found that robots boosted annual productivity growth by 0.36 percentage points over the 1993-2007 period.6 This is impressive because, if this estimate holds true for the U.S., robots' contribution to the 2½% average annual U.S. total productivity growth over the period was 14%. To put the importance of robotics into historical context, its contribution to productivity so far is roughly on par with that of the steam engine (Chart II-9). It falls well short of the 0.6 percentage point annual productivity contribution from the IT revolution. The implication is that, while the overall productivity performance has been dismal since 2007, it would have been even worse in the absence of robots. What does this mean for inflation? According to the "cost push" model of the inflation process, an increase in productivity of 0.36% that is not accompanied by associated wage gains would reduce unit labor costs (ULC) by the same amount. This should trim inflation if the cost savings are passed on to the end consumer, although by less than 0.36% because robots can only depress variable costs, not fixed costs. There indeed appears to be a slight negative relationship between robot density and unit labor costs at the industrial level in the U.S., although the relationship is loose at best (Chart II-10). Chart II-9GPT Contribution To Productivity Chart II-10U.S.: Unit Labor Costs Vs. Robot Density In theory, divergences in productivity across industries should only generate shifts in relative prices, and "cost push" inflation dynamics should only operate in the short term. Most economists believe that inflation is a purely monetary phenomenon in the long run, which means that central banks should be able to offset positive productivity shocks by lowering interest rates enough that aggregate demand keeps up with supply. Indeed, the Fed was successful in meeting the 2% inflation target on average from 2000 to 2007, when the impact of the IT revolution on productivity (and costs) was stronger than that of robot automation today. Also, note that inflation is currently low across the major advanced economies, irrespective of the level of robot intensity (Chart II-11). From this perspective, it is hard to see that robots should take much of the credit for today's low inflation backdrop. Chart II-11Inflation Vs. Robot Density 2. Human Displacement A key question is whether robots and humans are perfect substitutes. If new technologies introduced in the past were perfect substitutes, then it would have led to massive underemployment and all of the income in the economy would eventually have migrated to the owners of capital. The fact that average real household incomes have risen over time, and that there has been no secular upward trend in unemployment rates over the centuries, means that new technologies were at least partly complementary with labor (i.e., the jobs lost as a direct result of productivity gains were more than replaced in other areas of the economy over time). Rather than replacing workers, in many cases tech made humans more productive in their jobs. Rising productivity lifted income and thereby led to the creation of new jobs in other areas. The capital that workers bring to the production process - the skills, know-how and special talents - became more valuable as interaction with technology increased. Like today, there were concerns in the 1950s and 1960s that computerization would displace many types of jobs and lead to widespread idleness and falling household income. With hindsight, there was little to worry about. Some argue that this time is different. Futurists frequently assert that the pace of innovation is not just accelerating, it is accelerating 'exponentially'. Robots can now, or will soon be able to, replace humans in tasks that require cognitive skills. This means that they will be far less complementary to humans than in the past. The displacement effect could thus be much larger, especially given the impressive advances in artificial intelligence. However, Box II-1 discusses why the threat to workers posed by AI is also heavily overblown in the media. The CEP multi-country study cited above did not find a large displacement effect; robot usage did not affect the overall number of hours worked in the 23 countries studied (although it found distributional effects - see below). In other words, rather than suppressing overall labor input, robot usage has led to more output, higher productivity, more jobs and stronger wage and income growth. A report by the Economic Policy Institute (EPI)7 takes a broader look at automation, using productivity growth and capital spending as proxies. Automation is what occurs as the implementation of new technologies is incorporated along with new capital equipment or software to replace human labor in the workplace. If automation is increasing 'exponentially' and displacing workers on a broad scale, one would expect to see accelerating productivity growth, robust capital spending, and more violent shifts in occupational shares. Exactly the opposite has occurred. Indeed, the report demonstrates that occupational employment shifts were far slower in the 2000-2015 period than in any decade in the 1900s (Chart II-12). Box II-1 The Threat From AI Is Overblown Media coverage of AI/Deep Learning has established a consensus view that we believe is well off the mark. A recent Special Report from BCA's Technology Sector Strategy service dispels the myths surrounding AI.8 We believe the consensus, in conjunction with warnings from a variety of sources, is leading to predictions, policy discussions, and even career choices based on a flawed premise. It is worth noting that the most vocal proponents of AI as a threat to jobs and even humanity are not AI experts. At the root of this consensus is the false view that emerging AI technology is anything like true intelligence. Modern AI is not remotely comparable in function to a biological brain. Scientists have a limited understanding of how brains work, and it is unlikely that a poorly understood system can be modeled on a computer. The misconception of intelligence is amplified by headlines claiming an AI "taught itself" a particular task. No AI has ever "taught itself" anything: All AI results have come about after careful programming by often PhD-level experts, who then supplied the system with vast amounts of high quality data to train it. Often these systems have been iterated a number of times and we only hear of successes, not the failures. The need for careful preparation of the AI system and the requirement for high quality data limits the applicability of AI to specific classes of problems where the application justifies the investment in development and where sufficient high-quality data exists. There may be numerous such applications but doubtless many more where AI would not be suitable. Similarly, an AI system is highly adapted to a single problem, or type of problem, and becomes less useful when its application set is expanded. In other words, unlike a human whose abilities improve as they learn more things, an AI's performance on a particular task declines as it does more things. There is a popular misconception that increased computing power will somehow lead to ever improving AI. It is the algorithm which determines the outcome, not the computer performance: Increased computing power leads to faster results, not different results. Advanced computers might lead to more advanced algorithms, but it is pointless to speculate where that may lead: A spreadsheet from 2001 may work faster today but it still gives the same answer. In any event, it is worth noting that a tool ceases to be a tool when it starts having an opinion: there is little reason to develop a machine capable of cognition even if that were possible. Chart II-12U.S. Job Rotation Has Slowed The EPI report also notes that these indicators of automation increased rapidly in the late 1990s and early 2000s, a period that saw solid wage growth for American workers. These indicators weakened in the two periods of stagnant wage growth: from 1973 to 1995 and from 2002 to the present. Thus, there is no historical correlation between increases in automation and wage stagnation. Rather than automation, the report argues that it was China's entry into the global trading system that was largely responsible for the hollowing out of the U.S. manufacturing sector. We have also made this argument in previous research. The fact that the major advanced economies are all at, or close to, full employment supports the view that automation has not been an overwhelming headwind for job creation. Chart II-13 demonstrates that there has been no relationship between the change in robot density and the loss of manufacturing jobs since 1993. Japan is an interesting case study because it is on the leading edge of the problems associated with an aging population. Interestingly, despite a worsening labor shortage, robot density among Japanese firms is falling. Moreover, the Japanese data show that the industries that have a high robot usage tend to be more, not less, generous with wages than the robot laggard industries. Please see Appendix II-2 for more details. Chart II-13Global Manufacturing Jobs Vs. Robot Density The bottom line is that it does not appear that labor displacement related to automation has been responsible in any meaningful way for the lackluster average real income growth in the advanced economies since 2007. 3. Inequality That said, there is evidence suggesting that robots are having important distributional effects. The CEP study found that robot use has reduced hours for low-skilled and (to a lesser extent) middle-skilled workers relative to the highly skilled. This finding makes sense conceptually. Technological change can exacerbate inequality by either increasing the relative demand for skilled over unskilled workers (so-called "skill-biased" technological change), or by inducing companies to substitute machinery and other forms of physical capital for workers (so-called "capital-biased" technological change). The former affects the distribution of labor income, while the latter affects the share of income in GDP that labor receives. A Special Report appearing in this publication in 2014 focused on the relationship between technology and inequality.9 The report highlighted that much of the recent technological change has been skill-biased, which heavily favors workers with the talent and education to perform cognitively-demanding tasks, even as it reduces demand for workers with only rudimentary skills. Moreover, technological innovations and globalization increasingly allow the most talented individuals to market their skills to a much larger audience, thus bidding up their wages. The evidence suggests that faster productivity growth leads to higher average real wages and improved living standards, at least over reasonably long horizons. Nonetheless, technological change can, and in the future almost certainly will, increase income inequality. The poor will gain, but not as much as the rich. The fact that higher-income households tend to maintain a higher savings rate than low-income households means that the shift in the distribution of income toward the higher-income households will continue to modestly weigh on aggregate demand. Can the distribution effect be large enough to have a meaningful depressing impact on inflation? We believe that it has played some role in the lackluster recovery since the Great Recession, with the result that an extended period of underemployment has delivered a persistent deflationary impulse in the major developed economies. However, as discussed above, stimulative monetary policy has managed to overcome the impact of inequality and other headwinds on aggregate demand, and has returned the major countries roughly to full employment. Indeed, this year will be the first since 2007 that the G20 economies as a group will be operating slightly above a full employment level. Inflation should respond to excess demand conditions, irrespective of any ongoing demand headwind stemming from inequality. Conclusions Technological change has led to rising living standards over the decades. It did not lead to widespread joblessness and did not prevent central banks from meeting their inflation targets over time. The pessimists argue that this time is different because robots/AI have a much larger displacement effect. Perhaps it will be 20 years before we will know the answer. But our main point is that we have found no evidence that recent advances in robotics and AI, while very impressive, will be any different in their macro impact. There is little evidence that the modern economy is less capable in replacing the jobs lost to automation, although the nature of new technologies may be affecting the distribution of income more than in the past. Real incomes for the middle- and lower-income classes have been stagnant for some time, but this is partly due to productivity growth that is too low, not too high. Moreover, it is not at all clear that positive productivity shocks are disinflationary beyond the near term. The link between robot usage and unit labor costs over the past couple of decades is loose at best at the industry level, and is non-existent when looking across the major countries. The Fed was able to roughly meet its 2% inflation target in the 1990s and the first half of the 2000s, despite IT's impressive contribution to productivity growth during that period. For investors, this means that we cannot rely on automation to keep inflation depressed irrespective of how tight labor markets become. The global output gap will shift into positive territory this year for the first time since the Great Recession. Any resulting rise in inflation will come as a shock since the bond market has discounted continued low inflation for as far as the eye can see. We expect bond yields and implied volatility to rise this year, which may undermine risk assets in the second half. Mark McClellan Senior Vice President The Bank Credit Analyst Brian Piccioni Vice President Technology Sector Strategy Appendix II-1 Why Is Productivity So Low? A recent study by the OECD10 reveals that, while frontier firms are charging ahead, there is a widening gap between these firms and the laggards. The study analyzed firm-level data on labor productivity and total factor productivity for 24 countries. "Frontier" firms are defined to be those with productivity in the top 5%. These firms are 3-4 times as productive as the remaining 95%. The authors argue that the underlying cause of this yawning gap is that the diffusion rate of new technologies from the frontier firms to the laggards has slowed within industries. This could be due to rising barriers to entry, which has reduced contestability in markets. Curtailing the creative-destruction process means that there is less pressure to innovate. Barriers to entry may have increased because "...the importance of tacit knowledge as a source of competitive advantage for frontier firms may have risen if increasingly complex technologies were to increase the amount and sophistication of complementary investments required for technological adoption." 11 The bottom line is that aggregate productivity is low because the robust productivity gains for the tech-savvy frontier companies are offset by the long tail of firms that have been slow to adopt the latest technology. Indeed, business spending has been especially weak in this expansion. Chart II-14 highlights that the slowdown in U.S. productivity growth has mirrored that of the capital stock. Chart II-14U.S. Capex Shortfall Partly To Blame For Poor Productivity Appendix II-2 Japan - The Leading Edge Japan is an interesting case study because it is on the leading edge of the problems associated with an aging population. The popular press is full of stories of how robots are taking over. If the stories are to be believed, robots are the answer to the country's shrinking workforce. Robots now serve as helpers for the elderly, priests for weddings and funerals, concierges for hotels and even sexual partners (don't ask). Prime Minister Abe's government has launched a 5-year push to deepen the use of intelligent machines in manufacturing, supply chains, construction and health care. Indeed, Japan was the leader in robotics use for decades. Nonetheless, despite all the hype, Japan's stock of industrial robots has actually been eroding since the late 1990s (Chart II-4). Numerous surveys show that firms plan to use robots more in the future because of the difficulty in hiring humans. And there is huge potential: 90% of Japanese firms are small- and medium-sized (SME) and most are not currently using robots. Yet, there has been no wave of robot purchases as of 2016. One problem is the cost; most sophisticated robots are simply too expensive for SMEs to consider. This suggests that one cannot blame robots for Japan's lack of wage growth. The labor shortage has become so acute that there are examples of companies that have turned down sales due to insufficient manpower. Possible reasons why these companies do not offer higher wages to entice workers are beyond the scope of this report. But the fact that the stock of robots has been in decline since the late 1990s does not support the view that Japanese firms are using automation on a broad scale to avoid handing out pay hikes. Indeed, Chart II-15 highlights that wage deflation has been the greatest in industries that use almost no robots. Highly automated industries, such as Transportation Equipment and Electronics, have been among the most generous. This supports the view that the productivity afforded by increased robot usage encourages firms to pay their workers more. Looking ahead, it seems implausible that robots can replace all the retiring Japanese workers in the years to come. The workforce will shrink at an annual average pace of 0.33% between 2020 and 2030, according to the Japan Institute for Labour Policy and Training. Productivity growth would have to rise by the same amount to fully offset the dwindling number of workers. But that would require a surge in robot density of 4.1, assuming that each rise in robot density of one adds 0.08% to the level of productivity (Chart II-16). The level of robot sales would have to jump by a whopping 2½ times in the first year and continue to rise at the same pace each year thereafter to make this happen. Of course, the productivity afforded by new robots may accelerate in the coming years, but the point is that robot usage would likely have to rise astronomically to offset the impact of the shrinking population. Chart II-15Japan: Earnings Vs. Robot Density Chart II-16Japan: Where Is The Flood Of Robots? The implication is that, as long as the Japanese economy continues to grow above roughly 1%, the labor market will continue to tighten and wage rates will eventually begin to rise. 1 Please see Technology Sector Strategy Special Report "The Coming Robotics Revolution," dated May 16, 2017, available at tech.bcaresearch.com 2 Note that this includes only robots used in manufacturing industry, and thus excludes robots used in the service sector and households. However, robot usage in services is quite limited and those used in households do not add to GDP. 3 Note that ICT investment and capital stock data includes robots. 4 Please see BCA Global Investment Strategy Special Report "Weak Productivity Growth: Don't Blame The Statisticians," dated March 25, 2016, available at gis.bcaresearch.com 5 Centre for Economic and Business Research (January 2017): "The Impact of Automation." A Report for Redwood. In this report, robot density is defined to be the number of robots per million hours worked. 6 Graetz, G., and Michaels, G. (2015): "Robots At Work." CEP Discussion Paper No 1335. 7 Mishel, L., and Bivens, J. (2017): "The Zombie Robot Argument Lurches On," Economic Policy Institute. 8 Please see BCA Technology Sector Strategy Special Report "Bad Information - Why Misreporting Deep Learning Advances Is A Problem," dated January 9, 2018, available at tech.bcaresearch.com 9 Please see The Bank Credit Analyst, "Rage Against The Machines: Is Technology Exacerbating Inequality?" dated June 2014, available at bca.bcaresearch.com 10 OECD Productivity Working Papers, No. 05 (2016): "The Best Versus the Rest: The Global Productivity Slowdown, Divergence Across Firms and the Role of Public Policy." 11 Please refer to page 27.

Overweight A key beneficiary of a tight job market are managed health care providers who see lifts in both employer-sponsored health plans and newly-affordable individual and family health plans. With the unemployment rate touching new lows and small-business hiring plans hitting new highs (unemployment rate shown inverted, second panel), the direction for premium revenue for this niche health care sub-index is clearly higher. At the same time as the top line is moving higher, cost inflation has dramatically decelerated, driven by collapsing pharma price inflation (third panel). The implication is outsized earnings growth this year. The market has clearly taken notice, rewarding the S&P managed health care index with a premium valuation (bottom panel). While this is an improvement from the discount multiple of much of the past decade, it remains a far cry from previous cyclical highs. We think an exceptional earnings growth phase should make this valuation expansion durable; stay overweight. The ticker symbols for the stocks in this index are: BLBG: S5MANH - UNH, AET, ANTM, CI, HUM, CNC.