Burying the "Fed model"

By Buttonwood

IF YOU invested in equities in the 1990s, you were bound to hear, sooner or later, about the "Fed model". This, I should hasten to add, was not the official position of the Federal Reserve but the name given to a relationship found by three economists between Treasury bond yields and stockmarket valuations. Lower bond yields lead to higher price-earnings ratios or, if you invert the latter, lower earnings yields. Those who thought that equities were ridiculously overvalued in the late 1990s were told that they "just didn't get it".

The idea was fairly simple. The present value of a stock was its future cashflows, discounted at some rate that was derived from the bond market. The lower the discount rate, the higher the present value. The brilliant thing about this measure, from the bulls' point of view, was that it was based on the prospective earnings ratio. So you could forecast rapid earnings growth, thereby lowering the prospective p/e, and claim that the market was "cheap". At the level of individual stocks, the trick was even simpler - investment banks needed to show that a stock was cheap in order to sell it. So they simply persuaded analysts to forecast future earnings growth that was sufficiently high to make the prospective p/e look cheap.

If you look at the chart, you can see that the Fed model did appear to work for about 15 years, and then it didn't. The two ratios have gone their separate ways over the last decade; low bond yields have not meant higher stock valuations, but the reverse. Thanks to Rui Antunes at Societe Generale for the data on p/es by the way.

So what was happening? In 2003, Cliff Asness of the fund management group AQR, wrote a piece in the Journal of Portfolio Management called "Fight the Fed Model" to argue that the whole thing was a case of money illusion. Let us start with the theory. There are two potential explanations for the link between bond yields and earnings yields. The first relates to competing assets. Low bond yields (and low cash rates, which usually accompany them) cause investors to give up in disgust and opt for equities instead. The second is the discounted cashflow approach discussed above.

There are two levels of rates to consider here, nominal and real (after inflation). If nominal, but not real, yields fall, then it is becuase of lower inflation expectations. But what happens to profits growth in these circumstances? As Asness wrote

It is absolutely true that, all else being equal, a falling discount rate raises the current price. All else is not equal, though. If when inflation declines, future nominal cash flow from equities also falls, this can offset the effect of lower discount rates. Lower discount rates are applied to lower expected cash flows.

But what about lower real rates? the same reasoning applies. The easiest way to think of this is that real rates balance the demand for capital with its supply. If businesses are confident, they will want to invest; they will demand more capital and the real rate will rise. If businesses are cautious, as they are now, they will not want to invest; demand for capital will fall and so will real rates. In short, high real rates are associated with a strong economy and low real rates with a weak one. It is not obvious that the latter is good for equities.

But central banks can intervene to push real rates down, as they have at the moment. It seems easy to accept that there are moments when real rates are inappropriately high. Their level is associated, not with strong growth, but with systemic fears about the financial system. Then the central banks can, and should, force rates down. One would expect such crises to be temporary. But this is an unusually long crisis and perhaps it needs unusually long intervention. Fair enough. But it is again not clear why this should lead to higher share prices. By signalling that it will keep rates low until 2015, the Fed is also signalling that it expects the economy to be weak until then; investors should adjust their profit expectations accordingly.

Now there is also the danger that central banks may keep rates inappropriately low; that credit growth is accordingly strong and that this money goes not into investment but into speculation (see Thailand in the 1990s). Investors may feel that the central bank will respond by cutting rates if the market wobbles and that this gives them protection against the downside. Arguably, this was the case from 1987-2007 when we had a series of bubbles, ending in housing.

Indeed, housing is another case where one can see Fed model reasoning at work. Low rates make mortgages more affordable for homebuyers; that means more home buyers, who can pay more; and that translates into higher prices. But just as shares are the capitalised value of future profits, houses are the capitalised value of future rents. Their price is thus dependent on rental growth and thus wage growth. If low nominal/real rates imply low inflation or low economic growth, then nominal/real rents and wages will also grow slowly. Prices can reach a level from which they must collapse; US mortgage rates are now close to a low but house prices are well off their record high.

Now you might feel there is a flaw in this reasoning with regard to the ultimate risk-free asset, the index-linked bond. Lower real yields on this asset automatically translate into higher prices. If other assets are priced off the index-linked bond, then they should rise too. But the return on risky assets is equal to the yield on the risk-free asset, plus a risk premium. If real yields are low, as they are now, that means investors are risk-averse. They thus demand a higher risk premium for holding equities or houses; it is not necessarily good news for those assets at all.

So what actually happened in the 1980s and 1990s? You have to remember the starting point which was the high inflation and interest rates of the early 1980s. As they fell, the yields on equities and bonds naturally dropped in tandem. This temporary correlation was treated as a golden rule and became used to justify ever higher share prices. But at some point, low bond yields became not a plus point for equities but a warning sign about future growth. And the Fed model broke.