The economy is healing
We know the US economy has turned the corner. At some point, activity will be strong enough that the Fed will begin to raise short-term interest rates from their intensive-care-unit level of today. (Yes, the Fed has already begun to raise the discount rate, but this has been to force the major banks back into the commercial paper market instead of dealing solely with the government.)
what happens when the fed funds rate rises?
Even though the initial move may be months off, when is less important than how high the rate is likely to go and the effect the move will have on stocks and bonds. It’s not too soon to begin thinking about any of this.
two parts to this post
–The first will be what financial theory, such as it is, says about what should happen.
–The second will be an examination of the historical record of fed funds rate increases over the past twenty-five years.
Theory
fed funds rate behavior
One of the Fed’s jobs is to help carry out our highest-level national economic objective: maximum sustainable growth with low and stable inflation. “low and stable” means nothing much higher than 2%.
This gives us our first benchmark. Under normal conditions the fed funds rate, the price of overnight interbank deposits, will be slightly positive in real terms–about .5%-1.0% higher than the target inflation rate.
If the economy is running too hot, the Fed temporarily raises the rate, both to telegraph its concern and to raise the cost of borrowing, thus slowing the economy back down. When the economy is down in the dumps, on the other hand, the Fed drops the rate below inflation to try to pep activity back up.
Today, the rate is at about .25%, meaning the economy has been in a train wreck and is barely breathing.
Where is normal, then? Assuming inflation is under control, that is, 2% or less (and I think it is), the fed funds rate should be somewhere around 2.5%-3.0%. That means that one the Fed starts upping the rate, it won’t stop until it has tacked on 200 basis points, and possibly as many as 250.
Long rates won’t rise by as much, since this isn’t bond investors’ first rodeo and thus to some degree have already priced in some of the short-term interest rise. The extent of the yield curve flattening (meaning a smaller rise in long rates than in short) remains to be seen, but the ten- and thirty-year bond yields could easily rise by 100 bp.
the effect on stocks
Strictly speaking, there is no independent demand either for stocks or for bonds. This is because, to a great extent, the two asset classes are substitutes for one another. There is demand for the more general class of long-lived investment securities, which includes both stocks and bonds.
Why is this distinction important? If stocks and bonds are more or less substitutes, then anything that changes the price of bonds also tends to change, in the same direction, the price of stocks, and vice versa.
As interest rates go up, the price of bonds goes down. So rising interest rates should exert downward pressure on stocks as well.
For government bonds, that’s the end of the story. Not so for stocks, however.
The Fed only raises interest rates when economic activity–and thus corporate profits–are expanding as well. Rising profits tend to put upward pressure on stock prices, offsetting part or all of the negative force of rising interest rates. One can at least imagine circumstances where interest rates are rising slowly enough, or profits are growing fast enough, that stock prices are either stable or have a rising bias.
bond-stock equilibrium
One can also look at what the equilibrium relationship between stocks and bonds should be. This is usually done by comparing the interest yield on government bonds with the earnings yield on stocks. The earnings yield is typically calculated as the annual earnings per share of an index like the S&P 500 divided by the price of the index. It’s the inverse of the PE ratio.
If we assume that the 2010 earnings per share for the S&P 500 will be 85 and the index level is a bit below 1200, then the earnings yield is about 7.0%, which equates to a price earnings ratio of 1/.07, or 14.
Let’s say that as a result of the rise in fed funds to 2.75%, the ten-year bond yield increases to 5.0%.
The “right” proportion between a unit of yield in the bond market and in the stock market is a function of investor preferences and changes as they do.
If investors were indifferent to whether the earnings came from stocks or bonds (a big if, but more or less the relationship that has prevailed over the past twenty years), equilibrium would occur when the interest yield and the earnings yield were equal. A 5% long bond would imply a 20 times price earnings ratio (a 5% earnings yield) on the stock market.
Whatever the exact right number for today’s world may be, one can observe that a unit of earnings is available today much more cheaply than has historically been the case in the stock market vs. the bond market.
To sum up: increasing earnings give stocks some defensive power against rising fed funds and long-bond interest rates. Also, relative to one another, stocks are priced much more cheaply than government bonds–again arguably giving them some protection against rising rates/lower bond prices.
History
I’ve gone back as far as 1986 to begin looking for periods when the fed funds rate was rising.
I don’t think it makes any sense to look at prior instances. Monetary policy was conducted far differently in the Seventies, for one thing. And early Volcker years were marked by his determination to crush inflationary expectations, and a resulting fed funds rate that peaked at 20%. Although rates were much higher than now even in the mid-Eighties, the situation then was much nearer what we might call normal. And the Greenspan era was just about to begin.
There are only five instances of sustained fed funds rise to look at. That’s not a big number to be generalizing from, but, although we should keep this in mind, this is usually the case in dealing with major macroeconomic forces.
The numbers are as follows:
period——————–ff rate change——————–S&P 500 change
8/86-9/87 (13 mos) +1.425% +26%
2/89-12/88 (10 mos) +3.25% +11%
9/92-2/95 (29 mos) +3.00% +13%
11/98-5/00 (19 mos) +1.75% +25%
6/03-6/06 (36 mos) +4.25% +29%.
Observation #1: rates rise a lot
The clearest point to be made is that in all five instances, rates rose substantially once the Fed started. In two cases, 1987 and 2000 the rate rise was interrupted, presumably before the sequence of increases could be completed, by sharp stock market declines. The first instance was the Crash of 1987, the second the popping of the Internet Bubble.
The three cases where the Fed chairman was free to complete his work, rates rose at least 300 bp.
In the time the most like the present situation, and in which Ben Bernanke had the most influence, the interest rate rise was the greatest, at over 4%.
The shortest period of rising rates was ten months. The three instances where the rate rise wasn’t interrupted lasted an average of a bit over two years.
Observation #2: stocks went up during each rate rise period (but read below)
From looking at the chart above, it’s clear that in all five instances stock prices rose while the fed funds rate was going up. Three of the five periods are straightforward examples of this. But on two occasions, the situation isn’t quite so simple.
–Black Monday
While it’s technically true that the stock market in the US rose by 25% in 1986-87, Wall Street fell by 25% in October 1987, the month following the final fed funds rate increase. This decline wiped out all the gains the market had made during the prior 13 months.
On the other hand, I think it would be hard to say that the fed funds rate increases caused the market collapse. At that time, the US was in terrible economic shape. We had large budget and current account deficits. The dollar was declining. For about a year domestic and foreign institutional and private investors had ceased buying Treasury bonds–Washington was being kept afloat by injections of funds from foreign central banks. And many (including myself) think the trigger for Black Monday was a disastrous mistake in the then-fledgling stock index derivatives market.
Arguably we would have been better off if the Fed had raised rates faster than it did.
—the Internet Bubble
The period from late 1998 to mid-2000 is another unusual period. The fed funds rate was rising, but not by much nor very quickly. The S&P rose by 25% over this time, but were already beginning to decline by the second quarter of 2000. Wall Street subsequently fell by about a third, more than wiping out all the gains of the Internet Bubble period.
Even today, I find it hard to give a coherent analysis of the Internet period, other than to say that it was a speculative mania like the Nifty Fifty period of the early Seventies–but centered on what were called TMT (Technology, Media and Telecommunications) stocks.
A decade after the peak, Intel still trades at less than a third of its price in early 2000, and Microsoft is half. Smaller TMT companies are either out of business or trading at, say, 5%-10% of their peak prices.
Still, again it’s hard to say that the rise in the fed funds rate caused the market decline. It was more like a Road Runner cartoon, where Wile E. Coyote suddenly finds he’s been running on air–and plummets.
What to say about stocks, then?
Maybe the best way to summarize is to say that in all five instances of fed funds rate rises, anticipated profit growth was enough to keep stocks from falling, despite the downward pressure from falling bond prices. In one case, the Internet Bubble, forecast profits in a large sector of the market, TMT, proved illusory and the market fell. In a second, Wall Street was undone by policy failures in Washington.
There doesn’t seem to me to be much evidence to bet that stocks will fall as the fed funds rate goes up. If anything, the faster rates go up, the better for preserving stock gains. And there’s a good chance that stocks will continue to rise all the while short rates are ratcheting up.
After two bubbles, Internet and housing/finance, already bursting on Wall Street within ten years, it’s hard to think we’re in bubble territory again. That leaves government policy as a worry for the market. Although I’ve been conditioned through thirty years in the market to think that US politics never really matters for Wall Street, I still think this is a genuine concern. Politics aside, though, rising rates are less of a problem for stocks than is commonly thought–or than marketers for bond funds, which get really hurt in periods like this, are willing to admit.