The Asset Growth Effect and Future S&P 500 Returns

Equity preference by investors as seen through the asset growth effect is a strong predictor of future stock market returns. In examining the capital cycle of equity preference, we can see a very strong correlation with 10-year future returns in the S&P 500. The outlook at present suggests single-digit returns in the next decade.


Capital Cycles and the Asset Growth Effect 

High returns on invested capital (ROIC) invite capital and new investment, bidding up asset prices to euphoric valuations while eventually creating excess capacity, plunging economic earnings below the cost of capital, sending a market lurching forward to what Joseph Schumpeter called, the ”gale of creative destruction”, the "process of industrial mutation that continuously revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one", forcing capital to exit, reducing prices, destroying excess capacity in its wake and reducing competition, until profitability recovers. The number of years that a business is expected to yield economic earnings is known as the growth appreciation period (GAP) and is defined by what Warren Buffett has referred to as its “moat”, the set of competitive advantages that support economic earnings, defying the tendency of ROIC to sink below the cost of capital in a process of mean reversion. Not only can some businesses or indeed asset classes “defend” their profitability, but, it is also true that there can be impediments to capital exiting a market, preventing the emergence of profitability, leading to “value traps”. 

The capital cycle is a concept pursued in Edward Chancellor’s book, Capital Returns, a compilation of reports written by Marathon Asset Management’s portfolio managers, and it is driven by a phenomenon known as the “asset growth effect”: the negative relationship between asset growth rates and subsequent abnormal returns.

Eugene Fama and Ken French found in their 2006 paper “Profitability, Investment and Average Returns” that, ceteris paribus, a company that has high rates of investment to generate expected profitability will have lower contemporaneous free cash flows to investors than one with similar profits but lower investment rates. They also found that while there does not exist a direct way to measure future investment, recent asset growth is a reliable proxy for expected investment, allowing for us to measure the effect. 

Michael Cooper, Huseyin Gulen, and Michael Schill found that events associated with asset growth, such as acquisitions, public equity offerings, public debt offerings, and bank loan initiations tend to be followed by abnormally low returns whereas events associated with asset contraction, such as spinoffs, share repurchases, debt prepayments, and dividend initiations tend to be followed by abnormally high returns. Cooper, Gulen & Schill also find that asset growth is a strong predictor of returns than traditional value, size, and momentum factors. In another paper, Cooper, Gulen and Schill make a quite sensation discovery: 

Suppose that on June 30th of each year from 1968 to 2006 an investor sorted U.S. stocks into ten equal portfolios based on the past year’s percentage change in total corporate assets. If the investor bought the stocks with the highest past growth in assets, the mean annual portfolio return over the 39 year period would have been just 4%. If, alternatively, the investor bought the stocks with the lowest past growth in assets, the mean annual portfolio return would have been 26%. The investor in the low growth stocks would have achieved a 22% return premium by simply sorting stocks on one of the most available pieces of business data—the growth in the book value of assets.

The asset growth effect is the organizing principle which explains the historical divergence in returns between value and growth stocks.  Indeed, Yuhang Xie in his paper, “Interpreting the Value Effect through Q-Theory: An Empirical Investigation”, suggests that the value effect disappears altogether once you take into account the asset growth effect.

The asset growth effect is not just a United States phenomenon: Xi Li and Rodney N. Sullivan found robust evidence for the asset growth effect and its persistence across time by examining evidence from 23 countries across the world. 

The Hierarchy of Monies 

In A Treatise on Money, John Maynard Keynes says that the “money of account” is "that in which debts and prices and general purchasing power are expressed ...[while]... money itself [is] that by delivery of which debt contracts and price contracts are discharged".  The distinction between “money of account” (in the case of the United States, the dollar) and money arises because “money” takes many forms, giving rise to what Hyman Minsky referred to as a "hierarchy of monies"

Monetary systems are hierarchical and money represents a debt-relation, a promise to pay; these “promises” or IOUs only become money if and only if other individuals or institutions accept them -what we narrowly think of as money, cash, is created by governments and in a sense, by banks, “although”, Hyman Minsky notes, “in principle every unit can "create" money - the only problem for the creator being to get it "accepted". Money supply growth is a balance sheet phenomenon as money is both an asset and a liability -one person’s asset is another person’s liability-. 

The gradations of the money hierarchy are determined by the ease of acceptation, which itself is a question of the uncertainty of the promise, with “ultimate money”, the means of final payment, at the top of the pyramid. Deposits, for example, represent promises to pay cash on demand, they are a bank’s liability and a depositor’s asset. Simplistically, we can delineate this hierarchy with “money”, the means of final settlement at the top and credit (securities), claims on income and possible capital gains, at the bottom. We can split money into two: cash and deposits, and securities into two: bonds, and equities, creating a simple hierarchy governed by uncertainty and liquidity. Of course, for the depositor, deposits are “means of final settlement”, so demarcating what separates “money” from “credit” is more complicated than it first seems, but we will side-step this issue as it does not impact what follows. The important thing to remember is that money supply is a balance sheet phenomenon and that monetary systems are hierarchical and characterized by debt-relations where instruments of varying degrees of acceptation are traded. 

The hierarchy can be expressed not only in terms of financial instruments but in terms of financial institutions, who act as “market makers” -the government, banks, firms (non-bank) and households-, to borrow from L. Randall Wray, standing ready to buy or sell financial instruments and who hold an inventory, or stocks of financial instruments. For example, below is a stylized balance sheet showing the stocks of financial assets held by the household and nonprofit sector:

The hierarchy of market makers matches the hierarchy of financial instruments, and each market maker faces relative prices for money, prices which link financial instruments across the hierarchy. In a complex financial system such as we have today, financial intermediation  acts to facilitate financial transactions on behalf of their clients. 

At the aggregate level investors’ balance sheets hold stocks of the supply of financial assets. The flow of funds across these financial assets reduces or diminishes stocks of each individual financial asset. The preference for each is the demand for that financial asset. The supply of financial assets is simply the market value of all financial assets. So, if there are $10 work of stocks, $20 worth of bonds -I assume no meaningful so these are non-financial liabilities- and $40 in cash, the supply is $70 with 14.28% allocated to stocks, 28.57% to bonds and 57.14% to cash. Investors bid for those financial assets and their prices adjust to demand. If the demand for equities is high, the price or value of stocks goes up and the willingness of investors to hold cash or bonds declines, causing their “prices” to fall. An investor may dispose of bonds for cash to purchase equities, or use existing cash to buy equities, both being acts that change the mix of the supply of financial assets without changing the supply. In our little model, the investor may use their entire cash horde to bid for stocks, so that the mix changes to $50 of stocks, $20 of bonds and no cash. The relative prices are a kind of “exchange rate” defining the relationship between all three financial assets. When we disaggregate this model into the competing portfolios that constitute the market for financial assets, whose owners ask and offer prices for stocks of financial assets in order to buy or sell them. The price adjusts so that there is always a willing holder of stocks of financial assets. 

If there is a “prevailing bias” to borrow a term George Soros uses in The Alchemy of Finance, expressed by a desire to hold stocks, there will be too few stocks to satisfy the market and too many bonds. If everybody wants to ditch their bonds to buy stocks, the price of stocks will be bid up and owners of bonds will ask for less and less until each individual’s portfolio and the and the aggregate portfolio matches the asset allocation preference. Wealth is created or destroyed even before funds flow, through the process of bidding and asking: if in our “economy” someone bids $5 for a stock previously selling for $1, the owners of that stock experience a 500% rise in their financial wealth, without any actual movement of money. As you have probably noted, it is difficult to have a shortage of supply of non-cash financial assets -though there are limits-, but, as Philosophical Economics points out, it is possible to have a shortage of yield, bidding up the supply of stocks and bonds without resolving the problem of yield. So far, the supply has remained constant, changing merely in mix, but net financial assets can be created, but only by real market makers opr investors.

Investor’s Share of Equity and Future Stock Returns

We can then link asset growth effect and the financial flows and capital stocks of the economy to determine if there is a link between the “equity preference” of investors -rather than their financial intermediaries-, by which one means the trend of expansion or contraction in the ownership of corporate equities, and subsequent 10-year returns.  

I define the “equity preference” as “the share of the investor’s equity and credit assets allocated to equities”:= Investor’s Equity Assets/(Investor Equity Assets + Investor’s Liabilities). 

The Financial Accounts of the United States (Z.1) gives the balance sheets of each sector of the economy and allows us to determine the equity preference of investors in the United States. We can define the universe of investors as being Households and nonprofit organizations sector, Nonfinancial corporate business, Federal government, State and local governments and the Rest of the world. The relevant data can be found in FRED Graphic

Equity preference or the growth in equity assets in the aggregate investor’s portfolio, is a remarkably powerful predictor of 10-year future returns. Given where we are in terms of equity preference, the next decade is likely to deliver low single digit returns. How this happens is a mystery: what the end result will be is something we can be sure of with a great deal of confidence, but not absolute certainty. 

While exploring this approach to looking at the market, I discovered a blog post and a peer-reviewed article around the same theme, which I will urge you to read. David C. Yang and Fan Zhang introduce the idea of “household share of equity”, which differs with this approach in that it narrowly considers the view of the household and nonprofit sector and also, it includes mutual fund holdings in its calculation of equity holdings. Though the approach differs and where I measure performance against the S&P 500’s future returns, they use future excess returns on the stock market. I also discovered the anonymous and brilliant author of the blog, Philosophical Economics, who cheekily terms this, “the Single Greatest Predictor of Future Stock Market Returns”. I was led to this point by a contemplation of the asset growth effect, whereas the author of the Philosophical Economics blog approaches this issue as a matter of supply and demand, which itself can be said to form part of capital cycle analysis and the asset growth effect, though he does not directly reference these. His discussion of the key drivers of price change is highly intuitive.