The groundbreaking consumption capital asset pricing model (CCAPM) developed by Nobel laureate Robert Lucas, John Dewey Distinguished Service Professor in Economics at the University of Chicago, provides a framework linking asset prices to the consumption and savings decisions of investors. According to CCAPM, an investment decision is a tradeoff between future and current consumption. It is assumed that investors would want to smooth consumption across time, so the market price of any financial asset should reflect how investors feel about the tradeoff.
While theory clearly delineates the role consumption should play in determining investment and market prices, empirical efforts to document this link have largely failed. According to University of Chicago Graduate School of Business professor Tobias J. Moskowitz, “Prior research on the CCAPM has focused on the relationship between asset prices and contemporaneous aggregate consumption. Since aggregate consumption is dominated by households who do not own stocks, it is not surprising that we fail to find a link in the data. It also seems consumption doesn’t adjust immediately, so looking at longer horizons of consumption growth is useful.
” In the recent study “Long-Run Stockholder Consumption Risk and Asset Returns,” Moskowitz, Christopher J. Malloy of London Business School, and Annette Vissing-Jørgensen of Northwestern University developed a variation of CCAPM that focuses on the long-run consumption risk of stockholders.
While the stock market may fluctuate wildly, movement in aggregate consumption is very small. This disparity has led researchers to conclude that unless investors were incredibly sensitive to tiny movements in aggregate consumption, there would be no hope of explaining large movements in equity prices. This puzzle, dubbed the “equity premium puzzle,” has plagued empirical tests of the model.
According to the CCAPM, a stock’s risk is measured by the degree to which asset prices move in tandem with consumption. Riskier assets have high covariance with consumption and thus have higher returns in order to compensate investors for this additional risk. In other words, if an investor invests in a stock that will perform badly when that investor has very low consumption, this investment will be very risky and hence the investor should demand a premium (a lower price) for this investment. On the other hand, if another investment exists that provides high payoffs when the investor has low consumption, this investment provides a source of “consumption insurance” when the investor needs it the most. The investor should be willing to pay a high price for such an investment.
Moskowitz explains: “When do investors, on average, face high and low consumption? Usually when they suffer a shock to their wealth or income. In such cases, an investor will want to have some investments in his portfolio that offer big payoffs during these times so he can continue consuming at his normal rate. The investor will be willing to pay a high price for investments that protect him against these situations. Thinking intuitively about when many investors may be faced with such a scenario, we can attempt to identify times of high and low consumption. For example, the business cycle provides one set of possibilities. A stock that has poor returns during a recession is considered riskier than a stock that performs poorly during economic boom times, because in the latter scenario the investor is able to maintain his consumption level more easily.”
While this story is economically sound, there has been little success linking consumption with asset prices.
“Data on current consumption growth does not line up with data on current asset returns,” says Moskowitz. “Instead, asset prices today tend to predict consumption growth for the next few years. Moreover, it turns out the consumption of stockholders, who determine prices, matters even more. Thus, the CCAPM model works very well if you line up stockholders’ long-run consumption growth with current asset returns.”
The authors demonstrate that focusing on the long-run consumption growth of stockholders goes a long way toward answering the “equity premium puzzle.”
Focusing on long-run aggregate consumption data is a major step in understanding the movement of asset prices and linking these changes to macroeconomic fundamentals. “However, this step only gets you part of the way,” adds Moskowitz. “Answering the big macroeconomic questions requires taking a step back to look at micro-level data, focusing exclusively on the set of households who likely impact asset prices.”
Using micro-level data, the authors identified consumption patterns of stockholders distinctly from consumption patterns of nonstockholders. Looking only at the consumption growth of stockholders, the authors find a very strong fit with asset return data. For nonstockholder consumption, there is no relation to asset returns, as would be expected.
“When thinking about financial markets, you have to think about the underlying economics of the investors in those markets,” says Moskowitz. “Intuitively, investors care about putting food on the table and what’s going to help them determine their well-being in the future.”
Focusing on Stockholders
The variation of the CCAPM developed
by Moskowitz, Malloy, and
Vissing-Jørgensen is based on two previous findings. First,
the relationship between returns and consumption growth
can be strengthened by measuring consumption growth over
several periods into the future. Long-run consumption risk
reflects the idea that stock returns might contain information
about future consumption. An alternative explanation is that
consumption simply adjusts slowly to news in the economy
and it may take several quarters of data to see household
consumption responses. In either case, the authors find that
stocks with high long-run consumption risk perform poorly
at the same time investors receive news that their future
consumption will be low.
Second, stockholder consumption is more strongly related to asset prices than aggregate consumption. According to the CCAPM, security prices arise from the saving and consumption decisions of investors, but the consumption patterns of noninvestors should not relate to prices at all, since by definition noninvestors do not buy and sell securities.
“When we look at aggregate consumption in the U.S. economy, much of this consumption is driven by people who don’t participate in the stock market, so looking for a link doesn’t make much sense,” says Moskowitz.
Using data from the Consumer Expenditure Survey from 1982 to 2004, the authors separated data on stockholder consumption and nonstockholder consumption. Long-run consumption growth is measured over 12 quarters. Long-run consumption risk for an asset is measured as the relationship between its quarterly return and long-run consumption growth.
“In the past, researchers mixed data from both stockholder and nonstockholder groups,” says Moskowitz. “The models economists have developed apply to a given investor in the stock market, and therefore it makes sense to only look at investors. Our real contribution is to look at stockholders and test that relationship directly.”
Moskowitz adds, “We’re chipping away at a very big puzzle. It’s important to study the relationship between today’s prices, today’s consumption, and future consumption. Moreover, you want to look at consumption that is relevant for determining asset prices—the consumption patterns of stockholders. When you do so, you get a very strong relation to asset prices.”
Long-Run Consumption Risk
The authors also used the long-run stockholder consumption
model to try to explain the findings of Eugene Fama,
Robert R. McCormick Distinguished Service Professor of
Finance at the University of Chicago Graduate School of
Business and Kenneth French, Carl E. and Catherine M.
Heidt Professor of Finance at the Tuck School of Business
at Dartmouth University, and others that small stocks have
higher average returns (lower prices) than large stocks and
that value stocks (high E/P or book value to market value)
have higher average returns than growth stocks (low E/P
or book value to market value). They find that the long-run
consumption patterns of stockholders also line up with the
returns to small, value stocks and help explain the puzzling
returns associated with these investments.
Focusing on the long-run consumption properties of stockholders not only helps resolve the “equity premium puzzle,” but also helps explain the difference in average returns across stocks, particularly those related to the size and value/growth measures of a stock. This result advances the understanding of asset prices in the economy, as well as the variation in prices across assets. Value stocks are simply riskier to stockholders than growth stocks, and hence command a premium in order to induce investors to hold them. Likewise, small stocks provide a riskier stream of payoffs with respect to stockholders’ future consumption, and sell for a lower price today in order to induce investment.
Economics Matters
“If we are able to understand the relationship between
consumption and asset prices, the next step is to look at the
underlying shocks that affect both consumption and asset
returns,” says Moskowitz.
When aggregate shocks to the economy occur, stockholders feel a disproportionate amount of those shocks. Nonstockholders are unlikely to participate in capital markets, partly because they want to be insulated from those shocks and do not necessarily have the income to invest in the market, so they have very low-risk consumption. Given that stockholders are willing to take more chances, they will bear more of the economic risk; as a result, their consumption will have a closer relationship to stock market performance.
Linking the fundamentals of underlying economics and consumption patterns to asset returns is critical for any market. There are many different dynamics to how the distribution and potential outcomes for consumption for investors will show up in local markets.
“Emerging market economies will need to think about the risks that investors face in terms of consumption,” says Moskowitz. “At a broader level, we’re talking about the underlying economics of all these markets and how that determines the way financial assets behave. As we open more borders, we’ll have to start thinking collectively about the consumption risks faced by all investors, and how investing across these different markets smoothes or exacerbates these risks.”
“Long-Run Stockholder Consumption Risk and Asset Returns,”
Christopher J. Malloy, Tobias J. Moskowitz, Annette Vissing-Jørgensen.