更大一直更好

BIGGER’S BEEN BETTER It’s never been better to be a giant company in the U.S., QUARTERLY LETTER but will it stay that way?

2Q 2019 Ben Inker | Head of Asset Allocation If you will accept a little oversimplification, the last decade or so of global equity EXECUTIVE SUMMARY market performance can be summarized as follows. U.S. stocks have profoundly The largest companies in the U.S. have outperformed stocks in the rest of the world, whether other developed markets or seen marked improvement in their emerging markets. This outperformance has been partially driven by U.S. P/E ratios profitability on a number of measures expanding more than in other markets. But the largest driver of the outperformance over the last 30 or so years, while has been the massive superiority of earnings growth in the U.S. relative to anywhere smaller companies have seen very little change. The improvement for the largest else. This superior earnings growth has been driven not so much by strong top-line companies is probably due to increasing growth, but by expanding profitability by U.S. companies relative to sales, gross market power for these firms and a playing profits, or other measures that can plausibly be used as proxies for economic capital. field that has tilted in their favor. It is my Unlike in past cycles, this rising profitability seems to have been neither a result of, nor guess that the field may well be starting to a driver of, increased corporate investment. Digging a little deeper, we can see that tilt back away from them in a way that will the improvement in profitability has occurred only in the largest companies. These hurt their profitability in the future. But the companies have been out-earning their smaller brethren by increasing margins over safer bet is that that process will be slow the past 25 years or so. The long period of their improvement suggests this effect is not and uncertain. Given that our valuation something we should expect to correct over a single business cycle, but my guess is models assume margins revert relatively that the world in the future will be less favorable to these large, dominant companies quickly, we may well be a little too tough than is true of the current environment. If we were to adjust the assumptions of margin on these companies, but even giving them reversion in our forecasts to account for a slower pace of this reversion, our forecast for the full benefit of the doubt on their future U.S. large cap stocks (as well as high quality stocks) would improve noticeably, but the profitability would still leave them looking adjustment would still leave them looking meaningfully worse than the other groups of expensive versus the rest of the world. equities we forecast.

For a very long time, U.S. corporate profits seemed to be an exceptionally well-behaved mean-reverting series. Exhibit 1 shows corporate profits relative to GDP in the U.S. since 1929, with dotted lines marking the average level across the 20th century and the period 2004 to 2019.

EXHIBIT 1U.S. CORPORATE PROFITS/GDP 12% 10% 8% Profits/GDP 6% 4% 2% 0% -2% 1929 1937 1945 1953 1961 1969 1977 1985 1993 2001 2009 2017 As of March 2019 | Source: Federal Reserve Note: U.S. Corporate Profits after tax with IVA and CCadj.

Bigger’s Been Better | p2 Up through the 2001 recession, this series was remarkably mean-reverting, with the expected fluctuations across the business cycle and the only outlier occurring in the depths of the Great Depression. Sometime around 2004, however, this series experienced a decided upward shift and since then has averaged 9% of GDP versus the 6% average for the 20th century. Though it has maintained its cyclical variations, the series seems very clearly to have been varying around a different mean. This shift, as profound as it seems to have been, has not actually been that widespread among companies. Exhibit 2 shows a conceptually similar series, profits as percent of value-added for corporations.1 It isn’t exactly the same as profits/GDP, but it is available on an individual company basis.

EXHIBIT 2PROFITS/VALUE ADDED FOR U.S. COMPANIES BY MARKET RANK 35% 30% 25% 20% 15% 10% 5% 0% -5% 1986 1990 1994 1998 2002 2006 2010 2014 2018 Top 50 Next 450 Next 2500 As of 7/31/19 | Source: GMO, Worldscope What we immediately see from Exhibit 2 is that the increase in profitability has not been at all evenly distributed. With the number of lines on the chart, however, it’s probably easier to understand the magnitudes in table form.

TABLE 1: PROFIT/VALUE ADDED AVERAGES ACROSS PERIODS Top 50 Next 450 Next 2500 1986-1995 15% 13% 11% 1996-2005 20% 14% 10% 2006-2019 24% 18% 11% Improvement since 1986-1995 62% 37% 5% Value added for corporations is another term for gross As of 7/31/19 | Source: GMO, Worldscope profit – revenue less the cost of goods sold.

There is obviously no magic to picking the periods. These three periods have the advantage of each containing one The shift in fortunes for different companies clearly predates the shift in overall profits. recession, which helps keep the averages “fair.” Breaking down the period into the first decade, second decade, and last 13 years shows2 Bigger’s Been Better | p3 that the top 50 companies really started to distance themselves from the pack by the late 1990s and today have opened a huge gap on the rest of the listed universe. Smaller companies, by contrast, have seen little or no change over the period, which leaves the large majority of public corporations in the U.S. enjoying no benefit from the greatest advance in overall corporate profitability in U.S. history!

There was another interesting change in the profit picture as this shift was taking place. The relationship between profits and investment appears to have completely broken down, as we can see in Exhibit 3.

EXHIBIT 3U.S. CORPORATE PROFITS/GDP VS. NET BUSINESS INVESTMENT 12% 8% Net Domestic Business Investment/GDP Profits 7% 10% 6% 5% 8% 4% Profits/GDP 6% 3% 2% 4% Net Business 1% Investment 0% 2% -1% 0% -2% 1960 1967 1974 1981 1988 1995 2002 2009 2016 As of 3/31/19 | Source: Federal Reserve While net business investment has always been a volatile series, up until the early 2000s there used to be a meaningful relationship between profits and investment – high profits were associated with high rates of investment and low profits the opposite. The correlation between them was about 35%. But just at the time that the profit series made its lurch upward, net business investment made a lurch downward, from which it has not recovered.

On the face of it, this is extremely odd. If profitability is high, corporations should naturally respond by increasing investment. While in principle this tendency could be stopped by there being a particularly high cost of capital, it is pretty obvious this has not been the case over this period. Exhibit 4 shows the Shiller earnings yield of the S&P 500 (the inverse of the Shiller P/E). This may not be a perfect measure of the cost of equity capital for corporations, but it’s a pretty good one, and you certainly couldn’t describe the period since the early 2000s as being a high-cost environment for raising equity.

Bigger’s Been Better | p4 EXHIBIT 4: SHILLER EARNINGS YIELD 16% 14% 12% 10% 8% 6% 4% 2% 0% 1960 1967 1974 1981 1988 1995 2002 2009 2016 Source: Robert Shiller It has also not been an expensive time to raise capital through borrowing. Exhibit 5 shows the yield of Baa corporate debt, which again has been conspicuously low in this period.

EXHIBIT 5BAA CORPORATE BOND YIELD 20% 18% 16% 14% 12% 10% 8% “ 6% 4% 2% ...the change in corporate 0% business models has 1960 1967 1974 1981 1988 1995 2002 2009 2016 not been matched by As of June 2019 | Source: Moody’s via Federal Reserve Bank of St. Louis a change in financial accounting standards, One complaint that one could reasonably make about the data I have shown is that the so perhaps stated profits change in corporate business models has not been matched by a change in financial aren’t a good proxy for accounting standards, so perhaps stated profits aren’t a good proxy for what is truly what is truly going on. going on. Moving from accounting income to economic profit is a tricky and inexact business, but our Global Equity team has spent several years going meticulously through corporate accounts to try to properly account for the expenses that are really investments and adjust for inconsistent inventory accounting, among other corrections. While there is always more to be done, the team can now generate adjusted earnings figures that are almost unquestionably closer to economic reality. Exhibit 6 shows a version of the earlier chart on profitability by market rank.

Bigger’s Been Better | p5 EXHIBIT 6: ECONOMIC PROFITS/VALUE ADDED FOR U.S.

COMPANIES BY MARKET RANK 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 -0.05 1986 1990 1994 1998 2002 2006 2010 2014 2018 Top 50 Next 450 Next 2500 As of 7/31/19 | Source: GMO, Worldscope You would be forgiven for not noticing the difference between this and Exhibit 2, but the data is at least subtly different. Interestingly enough, when comparing the three periods this version of the data shows a slightly smaller advance for the largest companies and a small decrease in profitability in smaller cap firms, although the basic pattern is very similar.

TABLE 2ECONOMIC PROFIT/VALUE ADDED AVERAGES ACROSS PERIODS Top 50 Next 450 Next 2500 1986-1995 17% 16% 13% 1996-2005 22% 16% 13% “ 2006-2019 26% 19% 12% The gap between the Improvement since 1986-1995 50% 24% -8% improving fortunes for As of 7/31/19 | Source: GMO, Worldscope the largest firms and stagnant fortunes for the bulk of companies does The largest 50 companies have seen a 50% improvement in their adjusted profitability help explain the odd lack since the earliest decade, while the next 450 have seen a much smaller 24% increase and the rest an 8% decrease. Clearly, the upward shift in profitability is not driven by of relationship between archaic accounting standards, although it is interesting to see that properly accounting profits and investment. for R&D, advertising, and other expenditures that analysts argue GAAP gets wrong doesn’t particularly flatter recent years relative to the more distant past.3 But the gap between the improving fortunes for the largest firms and stagnant fortunes for the bulk of companies does help explain the odd lack of relationship between 3 profits and investment. The largest firms fund their investments more or less entirely Adjusting the accounting improves profitability for all the out of internal cash flow. The only notable exception to this is when they buy other groups in all periods, with no particular tendency for the recent years to be more affected than earlier ones. companies, and that, while certainly an investment from the firm’s standpoint, is not Bigger’s Been Better | p6 an investment from an overall economic perspective because assets are only shifted around, not created.4 The companies that do use outside money to make investments are generally small companies. Because the profitability of smaller companies is not particularly high, it cannot be much of a surprise that they feel no burning need to raise capital to ramp up their investments.

This still does leave the question as to why the largest firms are not using their enhanced cash flow to increase their own investment rates. After all, if these companies have increased their profits significantly, they could increase their investments out of internally generated cash flow. But the payout ratios of these companies, inclusive of the money they spend on stock buybacks, has instead grown substantially over the past 33 years, as we can see in Exhibit 7.

EXHIBIT 7EFFECTIVE PAYOUT RATIO 140% 120% 100% 80% 60% 40% 20% 0% -20% -40% -60% -80% 1986 1990 1994 1998 2002 2006 2010 2014 2018 Top 50 Next 450 Next 2500 As of 7/31/19 | Source: GMO, Worldscope Actually, given the gyrations of stock issuance by smaller companies and their occasional swoons in profitability, it’s a little hard to see the shifts displayed in the exhibit.5 Table 3 shows averages for the groups in the same periods as the profitability figures.

TABLE 3AGGREGATE PAYOUT AS PERCENT OF ECONOMIC PROFITS Top 50 Next 450 Next 2500 1986-1995 57% 41% -1% 1996-2005 55% 46% 0% 4 2006-2019 71% 72% 32% Buying the equity of another company does not create any new physical or intellectual capital as it is a purely financial transaction. There may be goodwill created, but that is an Change since 1986-1995 14% 31% 33% accounting concept, not really an economic one.

I took out the data from April 2009 to February 2010 for As of 7/31/19 | Source: GMO, Worldscope the “Next 2500” group as the combination of sometimes negative earnings and some significant dilution left the numbers not looking meaningful.

Bigger’s Been Better | p7 All groups have shown a significant increase in payout. In the case of larger companies, they are now paying out over 70% of their earnings to shareholders. Smaller cap companies pay out much less than their larger counterparts at just over 30% of earnings, but until fairly recently they paid out nothing net of equity issuance, making their shift over the period the largest of any group.

So why would we have a combination of enhanced profitability and decreasing investment? The most plausible answer seems to me to be that increased industry concentration and market power by the largest companies means the bulk of their profits takes the form of economic rents in the businesses in which they are dominant.6 Because the firms are already dominant in their spaces, there may not be a lot of expanding they can do short of branching out into new business lines, where their position will likely not be as dominant and their profit margins far less enticing.

The investment opportunities of these extraordinary companies do not seem to have grown on pace with the economic rents they are capturing in their current business.

And the rest of corporate America is not profitable enough to justify ramping up its investment. What does that tell us about the future of growth and profits in America?

The plus side of low rates of investment is that it means less chance of overcapacity, which pushes down profit margins, being created. The minus side is, of course, slow growth. While it is probably unfair to entirely blame the slowdown in U.S. productivity growth on the fall-off in corporate investment,7 the period since 2005 has been a uniquely dismal one for productivity growth in the U.S. We can see this in Exhibit 8.

EXHIBIT 85-YEAR MOVING AVERAGE PRODUCTIVITY 4% 3% 1951-2005 2% While this seems an extremely logical reason, I have to admit 1% that our attempts to “prove” that the shift in profitability has been driven by an increase in industry concentration, either cross-sectionally (more concentrated industries being more profitable) or longitudinally (industries where concentration 0% increased the most improved their profitability) have not 1951 1958 1965 1972 1979 1986 1993 2000 2007 2014 panned out. This could easily be because of the relative As of 3/31/19 | Source: Federal Reserve crudeness of the concentration data we have, or the simplicity of the Herfindahl-Hirschman Indexes we use to define concentration.

The most striking shift in productivity growth in recent While productivity has risen from the truly dismal rates in the immediate aftermath years has been the huge fall-off in growth in the IT sector. of the Global Financial Crisis, productivity growth is half that of the first 55 years as From 1998 to 2005, productivity growth in computer and shown in Exhibit 8 and a third of what we saw in the glory days of the late 1990s and electronic products was 12.9% per year according to the early 2000s. It’s hard to see what would prompt a strong recovery in productivity in the Bureau of Economic Analysis (BEA). Since then it has been 3.9% per year. The demise of Moore’s Law, Denard Scaling, absence of a strong recovery in corporate investment, and there aren’t a lot of reasons Kryder’s Law, and other measures of advances in computing to expect a strong recovery in corporate investment. The largest companies presumably power surely played an important role in the slowdown apart quite like the world as it is, given their place in it. While that should not stop them from from any impact from falling investment. But across the making investments where they foresee a high return on capital, they have a lot to lose 63 industries that the BEA tracks the median productivity slowdown has been 2.2%, so it certainly isn’t the case that from change and are likely to actively work against disruptive progress. Even apart everything is caused by technological roadblocks. from a desire to maintain the status quo, dominant companies may well not be in a Bigger’s Been Better | p8 position to want to aggressively grow for fear of attracting the attention of regulators in the U.S. and abroad. That attention is certainly more clearly focused on the largest technology firms today, although to date the fines and business restrictions placed on them have not put much of a dent in their business models.

As to the question of what this suggests for the future of the U.S. stock market, it’s hard to know for sure, but a few points seem worth making. First, if the shift upward in the profitability of the largest companies has been going on over multiple business cycles, we shouldn’t expect it to unravel the same way that prior cyclical peaks in profitability did. That is not to say that there is not a cyclical element to profitability today, but rather that the cyclical factor looks to be superimposed on another longer-term shift.

Our equity forecasts split the difference between assuming profitability is stable over time and that enduring shifts can happen, but they are biased toward an assumption of stability, with two-thirds of the weight on models that assume profitability is stable in “ the long run.

Because the profitability shift for the largest companies in the U.S. is unprecedented, ...no company’s it’s hard to look back at history to determine how we expect the future should evolve for them. The trend has certainly been in their favor and absent any change in the competitive position environment it is tempting to assume things will continue to get better for them. There is safe forever. Today’s are a couple of potential flaws in that reasoning. The first is the fact that no company’s champions have competitive position is safe forever. Today’s champions have certainly done well historically – they wouldn’t be the largest companies in the U.S. if they hadn’t done certainly done well well, after all. But the history of dominant companies shows that their extraordinary historically. But the profitability tends to decay over time. That decay may be slow or rapid depending history of dominant on the company. For the most stably dominant companies, such as the ones favored companies shows that by our Focused Equity team in the GMO Quality Strategy, history suggests that their profitability decays to normal over something like a 30-year period. Now, 30 years is a their extraordinary long time to benefit from above-average profitability, justifying a significant valuation profitability tends to premium over average companies. But from the standpoint of forecasting the future decay over time. profitability for such companies, it argues for a slow reversion rate for profitability, not no reversion.

But while that might be the right way to think about the largest companies individually, a group of companies needn’t act like an individual company. The largest stocks in the S&P 500 are a different group of companies than they were in 1996, or 2006, and yet their aggregate profitability has risen even as some former stars have seen their fortunes fade. On this front, the question is really whether the environment that has favored these dominant firms is likely to remain so biased in their favor. My guess is that it is not. The recent announcement of anti-trust investigations against the largest technology and related communications firms seems unlikely to be a one-off. Apart from the simple issue of their growing dominance and staggering profits, the negative societal and privacy impacts of their basic business models make it harder to paint them as plucky upstarts improving the world and easier for critics to characterize them as disturbing big brother entities feeding off the worst instincts of humanity. Health care companies face a similarly daunting level of unpopularity, driven by their pricing See Naidu, Posner, and Weyl “Antitrust Remedies for Labor Market Power,” https://papers.ssrn.com/sol3/papers. practices combined with the striking disconnect between the amount the U.S. spends cfm?abstract_id=3129221; Mainescu and Hovenkamp on health care versus the rest of the world and the health outcomes that spending seems “Anticompetitive Mergers in Labor Markets,” https:// to deliver. The growing interest in academia and beyond in studying the deleterious papers.ssrn.com/sol3/papers.cfm?abstract_id=3124483; effects of large firms beyond looking at the direct impact on consumer prices also and Ashenfelter, Hosken, and Weinberg “Did Robert Bork Understate the Competitive Impact of Mergers? Evidence suggests a broader push against dominant firms on principle is increasingly likely.8 from Consummated Mergers,” https://www.journals. uchicago.edu/doi/abs/10.1086/675862?journalCode=jle# How long these forces will take to play out and how negative they will be for the largest among others looking at moving antitrust theory beyond the firms is hard to know with any certainty. But it does seem to me that the wiser bet is to Bork Doctrine and in a somewhat more assertive direction.

Bigger’s Been Better | p9 Ben Inker assume the world will not remain entirely safe for dominant companies. If we took the Mr. Inker is head of stance that the next seven years would see a one-third reversion of the trend of large GMO’s Asset Allocation company margins back to their long-term average, our forecast for the S&P 500 would team and a member improve by 1.4%.9 Given that our current forecast for that group is -3.8% and -0.9% on of the GMO Board of our two base scenarios, that assumption would raise the forecast to -2.4% and +0.5%.

Directors. He joined GMO That is enough to shrink the U.S. “margin of inferiority” relative to other equities, but in 1992 following the completion of his B.A. in still leaves it at the bottom of the equity barrel. Even an assumption that the average Economics from Yale University. In his years profitability did not revert at all would not make a profound difference to the forecast, at GMO, Mr. Inker has served as an analyst for as it would increase the forecast by about 2%, again not enough to lift it out of the the Quantitative Equity and Asset Allocation cellar in the equity rankings or to meaningfully change the desired holdings for the teams, as a portfolio manager of several equity portfolios we run.10 To see why, Exhibit 9 shows the forecast for the S&P 500 with and and asset allocation portfolios, as co-head of without the potential changes relative to some of our favorite equity assets.

International Quantitative Equities, and as CIO of Quantitative Developed Equities. He is a CFA charterholder.

EXHIBIT 9S&P 500 VS. OTHER EQUITY GROUPS Disclaimer 12% 11.1% Annual Real Return Over 7 Years The views expressed are the views of Ben Inker 10% 8.8% 8.2% through the period ending September 2019, 8% 6.8% 5.6% and are subject to change at any time based 6% 4.4% 4.8% on market and other conditions. This is not an 4% 3.1% 2.7% offer or solicitation for the purchase or sale 2% 0.5% 1.1% 0.7% of any security and should not be construed 0% as such. References to specific securities -2% -0.9% and issuers are for illustrative purposes only -2.4% -1.8% -4% and are not intended to be, and should not be -3.8% -6% interpreted as, recommendations to purchase S&P 500 S&P 500 S&P 500 U.S. Int'l Int'l EM EM or sell such securities. with 1/3 with no Small Value Value Profit Profit Value Copyright © 2019 by GMO LLC. Reversion Reversion All rights reserved. Mean Reversion Partial Mean Reversion As of 6/30/19 | Source: GMO The groups we are excited about within global equities – U.S. small value, international 9 developed value, and, above all, emerging markets value – are all at least at a 4.5% The improvement comes from the fact that many of higher forecast return in the friendliest version of the forecast for the S&P 500, with our models currently assume that all their abnormal EM value a stunning 10% to 10.6% higher.11 profitability will disappear over seven years rather than a longer period, so this slows down the reversion. For the moment, we are not making any changes to our “official” asset class forecasts Shifting this to one-third reversion effectively treats them despite a guess that we are being a little tough on the U.S. mega caps. Our preference collectively as a high-quality company.

10 is to not make one-off adjustments to our forecasts but look for the broader principle This would still leave us making a cyclical adjustment to behind an issue and find a way to systematically build it into our models. This allows us their profitability, but only assuming profitability reverted to to test whether a change improves the models across time and region or asset class, as the average of the 2006-2019 period. It’s just not in me to be well as making sure that as circumstances change we don’t find ourselves asking every prepared to assume that these companies are immune to even cyclical fluctuations to their profitability. month whether a one-off adjustment is still warranted or should change in magnitude. 11 But a challenge to our models is an opportunity to improve them, and we are Our naming conventions for our scenarios are a little confusing researching the general question of how to tease out cyclical aspects of profitability from here. The “Mean Reversion” scenario assumes that valuations potential secular changes. Should we find a method that genuinely seems to improve revert all the way to historical levels, while “Partial Mean Reversion” assumes that valuations can stay high versus history our understanding of asset class valuations, we will adjust our models accordingly. because of lower yields on fixed income. That is a different mean reversion than the mean reversion of profitability I have been talking about for U.S. mega caps.