Four things that make asset prices rise
I’ve been developing this particular piece for a while, but as the market starts to feel ready for a correction, I think it would be good to get this version out the door quickly. On the other hand, I’ve also been expecting some kind of a correction for some time, and yet it hasn’t happened, so perhaps there is no great rush after all.
The recent run-up in equities has had me scratching my head. As I said, I’ve been expecting some kind of market correction for some time now (as have many others), and yet it seems that the major equity indices march incessantly upwards. Why?
Yes, we are out of the economic free-fall of a year ago, and that surely justifies some improvements in price, but it is still very difficult to see where the next motor of US economic growth will be coming from, and therefore hard to see what cash flows are going to support higher equity prices. There is talk of another bubble in risky assets and particularly in equities, and it is legitimate to fear the effect of a new asset bubble popping so soon after the last. One must wonder what the effect will be on shaky banks, balance sheets, credit support, business and consumer confidence if a second round of liquidation hits us so soon after the last.
In moments like these, it is helpful to go back to the basics of finance theory and get a better feel for what kinds of things drive up asset prices. Here is a list in my mind, and I will go through the current economic context after listing the dynamics.
“Other things equal, ” asset prices rise when
- the asset’s expected return goes up
- the asset’s perceived risk goes down
- the asset’s correlation with other assets drops
- market “demands” for sources of excess return increase
1) When the expected return goes up, prices go up. Most people understand this part: if investors start to expect more return from an asset, and nothing else has changed, people will start to pay more for it until its price is in line with the new set of expectations. What people do not always realize is that when expected returns rise, and asset prices follow, the expected return will start to drop as time goes on. Mathematically, this is because the same additional returns (in dollar terms) will now be measured off of a higher starting price, and represent a lower return in percentage terms. How far will the return drop? Financial theory says that as the asset’s price rises, the expected return (in percentage terms) will continue to drop until it reaches the level consistent with what the market pays for that asset’s level of risk. Remember that, in this scenario, we are assuming that the riskiness of an asset has not changed, and that only the level of anticipated return has changed (presumably due to some previously unexpected event or piece of news).
2) When the perceived risk of an asset diminishes, prices go up. If upside risk and downside risk drop symmetrically (i.e. standard deviation goes down), the asset’s price will go up. In fundamental terms, one can interpret this rise as the effect of reduced “volatility drag”. There may also be an additional dynamic whereby investors who had previously perceived an asset as too risky to invest may now begin to invest in it, driving the price up further (an example would be a jump in a bond’s price when it crosses the “investment grade” threshold)
If risk declines asymmetrically - i.e. the chance of downside “surprises” diminishes more than the chance of upside surprises, then part of the price increase will be due to a change in expected return (since if the probability of a low return diminishes more than probability of an abnormally high return, the expected return will increase) and part of the increase will be due to the reduction in risk.
3) When an asset’s correlation with other assets declines, prices may go up. There are more assumptions needed to make this factor work, but the idea is that an asset’s diversification advantage in a portfolio will increase as correlation declines, therefore attracting additional investors and driving up the price. The assumptions required are that: a) the asset still has a positive expected return, b) that expected return is higher than the risk free rate, and c) the decline in correlation is not outweighed by a similar decline in expected return.
4) When demand for sources of return increases - regardless of fundamentals - prices go up. In other words, if the average investor wants or needs some kind of [excess] return so much that they are willing accept higher levels of risk to get it, asset prices will rise. In the fixed income world, this shows up as tightened credit spreads. In the world of equities, this appears as a shrinking equity risk premium. Risk premia - be it equity or credit - get smaller as the demand for returns increases because at every given level of risk, there are more investors willing to take it, and so the required compensation for that level risk is reduced. Interestingly, as prices rise and recently observed returns look impressive, future returns will start to decline to reflect the new risk premia. Another way to look at this is to remember that the willingness to take on more risk does not automatically guarantee that the economy can produce more return. The end result of this dynamic is that the risks from economic and fundamental factors should stay approximately the same (more on why they might not later), but the resulting returns will be lower because they are now measured off of a higher starting price (valuation).
Factors and the Present Economic Situation
Clearly, the emergence of an active government willing to use fiscal stimulus in early 2009 helped both to increase expected returns and to reduce the perception of fundamental risk. In short, financial armageddon had been a very real and frightening possibility, and by late February 2009 this possibility was now apparently off the table - or at least delayed substantially. The removal of the worst-case scenario from investors minds thus improved both Factor 1 (increased expected returns, now that large-scale bankruptcies seemed off the table) and Factor 2 (reduced downside risk, now that government guarantees and funds had been allocated), and both factors can account for the sharp rise in the S&P 500 index until probably about mid-year, when the P/E ratio based on 10 year real trailing operating earnings (sourced from Robert Shiller’s data) hovered around its historical average.
But the index did not settle at its historical average. Indeed, as I write, Shiller’s P/E calculation is 20.6 (for 1 Jan 2010), which is on the high side by historic standards, but actually rather low compared to most of the last decade. From 2003 to the onset of the credit crunch in 2007, this P/E measure tended to float around 25, but this was also at a time when fundamental risks appeared very low and the economy appeared to be firing on all cylinders. It may be that market participants are trying imagining themselves as satisfied at this level, which would suggest an S&P price in the 1375-1400 range.
Given the increased riskiness of the economy in general and the uncertain sources for corporate revenue growth, it is hard to justify P/E levels comparable to those in the 2003-2005 period. The expectations for rapid growth simply aren’t there, and the risks are increased on top of that. So what has been driving prices upwards?
Factor 3 may have some role in the upswing. During the dark months between September 2008 and March 2009, correlations among asset returns greatly increased, turning the investment world into essentially two assets: “safe” US Treasuries, and all other assets moving together. This became known as the risk-on/risk-off trade, and the increase in correlation (driven to some extent by selling any liquid assets to cover illiquid losses) itself drove prices downwards. As prices began to rise in 2009, separate assets stopped being so highly correlated and began to take on lives of their own. Although correlations are still fairly high across assets, the gradual de-coupling of asset returns has made diversification more effective and may be prompting a small increase in overall prices. This may reflect more the allocation shifts away from cash-type assets rather than the true attractiveness of risky assets.
The more important aspect that may be in play here is Factor 4 - a shrinking risk premium. Remember that the shrinking of the risk premium means an increased appetite for risk, also known as “reaching for yield.” It may well be that investors lost so much of their perceived wealth in the crash that they are unusually-willing to accept risk in the hopes of recovering their losses. Many may have engaged in a game of “double-or-nothing,” in which they are so far down that they believe the risk of missing out on a recovery is greater than the risk of losing to a subsequent market correction, or that being conservative and cautious will result in an outcome that is little better than losing substantially more by taking more risk. This is a behavioral phenomenon, one that might be better observed through technical charting. As long as the crowd is basically optimistic about price trends and is desperate for returns, prices will rise, but eventually the crowd will be prone to spooking. In this environment it makes sense to be hedged with stop-losses or options.
In short, the early period of the S&P’s rise appears to reflect the elimination of the doomsday scenario we collectively faced a year ago. However, since autumn 2009, the rise in the market appears to be more related to diminishing correlations and to increased risk appetite, which - while understandable - may not be economically rational. When the primary risk switches from “protecting capital” to “missing the recovery” it may be a sign that investors are reaching for yield and that markets are becoming highly “spookable.”