Tuesday, May 12, 2015

What equity market bubble?

Ruchir Sharma argues in today's WSJ ("The Federal Reserve Asset Bubble Machine") that the Fed's easy money policies have created dangerous asset market bubbles in "stocks ... bonds, houses and every other financial asset." In particular, he claims that his "composite valuation for the three major financial assets in America—stocks, bonds and houses—is currently well above levels reached during the bubbles of 2000 and 2007."

Although I don't agree that stocks and houses are egregiously overvalued, he is correct to point out that Fed policy is in uncharted waters: "we have analyzed data going back two centuries and found that until the past decade no major central bank had ever before set short-term interest rates at zero, even in periods of deflation." It's also the case that no central bank has ever expanded its balance sheet to the extent the Fed has, and excess reserves in the U.S. banking system are orders of magnitude higher (currently about $2.5 trillion) than they have ever been before. He is also correct when he concedes that it is hard to prove that Fed policy has been inflationary since measured inflation remains quite low.

His beef is with asset market prices, something the Fed has never attempted to control, but which he thinks they should pay attention to, if not worry about.

I do think that bond yields are low enough—relative to core inflation—to be considered "bubbly," but as I've noted for some time now, stocks appear to be trading only modestly above what looks to be fair value, and housing prices are still well below their 2006 highs. Furthermore, if money were so easy that as to be artificially inflating asset prices, I should think the dollar would be very weak, and certainly not trading—as it is—at historically average prices relative to other currencies. On balance, I don't think it's at all obvious that we are in the presence of dangerously inflating asset bubbles in stocks and housing. We may well get there, if the Fed fails to tighten monetary policy in a timely and sufficient manner, but we're not there yet.

What follows are some charts that help back up my assertions:

10-yr Treasury yields—the benchmark for most bonds—are indeed very close to their all-time lows.

But looking at nominal yields in a vacuum can be misleading. It's best to compare nominal yields to inflation, since that gives a much better idea of the true "valuation" of bonds. As the chart above shows, when core CPI inflation is subtracted from 10-yr Treasury yields they are still more than 100 bps above their modern-era lows. But abstracting from the period 2011-2013, real yields haven't been as low as they are today for the past 40 years. I think this qualifies as "bubbly," but to be fair it's also symptomatic of significant risk aversion (i.e., the world is willing to pay very high prices for the safety of Treasuries).

Real home prices are still 25% below their 2006 peak. Meanwhile, 30-yr fixed mortgage rates today are just under 4%, while they were 6% or higher in 2006. Housing is therefore much more affordable today than it was in 2006 or 2007. It could take years for housing to return to the valuations of 2006/2007.

At 18.5, the current PE ratio of the S&P 500 is only moderately higher than its 55-year average. It was far higher in the late 1990s. This could arguably be grounds for thinking that stocks are overvalued.

However, as the chart above shows, after-tax corporate profits have been running at record levels relative to GDP for the past several years. Profit margins, in other words, are at record levels and have been for some time now. Why shouldn't PE ratios be trading a bit rich considering how strong profits have been?

The chart above compares the earnings yield on stocks to the inverse of the real yield on 5-yr TIPS (which I use as a proxy for their price). Here we see that TIPS are still trading at very high prices, which means that the market is willing to pay a huge premium for the double safety of TIPS (which are default-free and protected against inflation). Earnings yields are also relatively high, which means the market is not very confident that profit margins will continue at current lofty levels. Together, these are symptomatic of a market that does not believe the economy is going to be very strong for the foreseeable future. In other words, this is a market that is still quite risk-averse—not a market that is over-priced.

As the chart above shows, the difference between the earnings yield on stocks and the yield on 10-yr Treasuries (a measure of the equity risk premium) is quite a bit higher than its historic average. If we grant that yields on Treasuries are "bubbly" and should be closer to 4% than to 2%, then the equity risk premium today would be about normal. So in this sense stocks are not over-priced at all; they are priced to the expectation that yields—and discount rates—will be much higher in the future.

The chart above compares the ratio of the S&P 500 to nominal GDP (blue line) to the yield on 10-yr Treasuries. The two tend to track each other, which is reasonable, since profits are a function of nominal GDP and the value of stocks is the present value of future profits discounted by some interest rate. Higher yields go hand in hand with lower valuations, and vice-versa. By this measure, stocks today have about the same valuation as they did in the 1960s, when bond yields were around 4%. In other words, stocks are not being discounted at current interest rates; rather, they are behaving as if interest rates were 4% instead of 2%. This is yet another sign that stocks are not wildly over-valued. (Note how stocks were indeed very overvalued by this measure in the late 1990s.)

If there is one fundamental reason for why the Sharmas of the world worry about over-priced asset markets but I don't, it is because they fail to understand that the Fed hasn't been super-easy. They start by assuming that monetary policy and zero interest rates are highly stimulative, and have been for years, and that therefore asset markets must be overpriced. They believe that monetary inflation can be found in asset prices, and it will eventually show up in the official inflation numbers.

In contrast, I start with the assumption that the Fed has not been stimulative. The Fed has merely accommodated a very strong demand for money, which in turn is a by-product of the huge degree of risk aversion that has been in evidence around world in the wake of the 2008 financial crisis. The Fed hasn't printed massive amounts of money, they've only swapped bank reserves for notes and bonds, in an effort to satisfy a risk-averse world.

I've explained this in greater detail in many posts over the past several years, and you can find some here. Quantitative Easing, huge excess reserves, low inflation, a relatively strong dollar and risk-averse markets all fit together, once you understand the Fed hasn't been super-easy. The equity and housing markets are not cheap, but they are not necessarily overpriced: in fact, I think they're still somewhat cautiously priced.


Benjamin Cole said...

Fascinating commentary by Scott Grannis.

It is an interesting School of Economics that holds that the free enterprise system is so feeble and unstable that low interest rates will cause massive and dangerous misallocation of capital and bubbles. Thus the Fed must ever keep interest rates high. Investors and business people lose all perspective when interest rates are too low.

I do disagree with Scott Grannis on one item: the Fed in fact printed money during QE. The Fed placed reserves into the commercial bank accounts of primary dealers, but the primary dealers paid cold hard (digitized) cash for the bonds they purchased.

Fine by me---I just wish the Fed had been even more aggressive.

steve said...

scott, your piece presupposes that by quantifying risk in stocks you might at some time sell to avoid losses when they become "frothy". otherwise, why bother with the exercise at all? with the greatest respect (yours is the only economic blog I have bookmarked!) I think this is just silly beyond extreme. there is virtually ZERO evidence that anyone has EVER been able to "time" stocks favorably (better than buy/hold) over the long run. therefore, all efforts to quantify risk are by definition, futile.

steve said...

forgot one thing. I've been in the business for 35 years and I cannot remember anytime in the past when there are so many negative pieces on why stocks must fall X %. very qualitative but still...

rkevanm said...

Thanks again. I have appreciated your explanation regarding QE, particularly 'transmogrification'. My feeling is also that US stock prices are not much overvalued, though it is just that--a feeling. So, are we to view indicators such as SP500 to regression trend, Total Market Cap to GNP, Household percentage of financial assets allocated to equities, Median P/E Multiple, and Median Price to cash flow as mostly just long-leading, or even very long-leading?

Anonymous said...

The negative case is fairly simple. First of all, the Wall St Journal 5/8/15 has the 500 at a 21 PE on trailing 12 months. That is high. Low interest rates provide a “carry” for margined stocks. People buy payments, not prices of houses. If rates go up significantly the carry is lost in margin accounts and monthly payments for homes goes higher.

That said, I think a high PE is justified not because of rapid growth of EPS but that with bond yields so low people are willing to look farther out into the future for their rate of return with stocks. (This is a more simple explanation than risk premiums and inflation expectations).

I don’t think rates will go up much if they even do go up. I do think we can be in a new normal of low rates. I think stock buybacks will buoy the stock market. And the economy could very well continue to grow slowly in spite of the recent slowdown that is of concern right now. An increase in rates just has to be discounted in stock prices by now. Sell on the rumor and buy on the news (after a dip when the first rate increase is announced).

Scott Grannis said...

steve: I appreciate your comments regarding the futility of trying to "time" the market. I am a long-term investor and make no attempts to get in and out of the market. Nevertheless, I do think it's valuable to have a sense for whether the market is too optimistic or too pessimistic; a general sense of valuations. That helps in asset allocation decisions, in how aggressive to be with one's positions, and how to deploy cash and make decisions on the margin.

Benjamin Cole said...

Retail sales stagnant.

William said...

Contrarian's Conundrum:

1) Investor’s Intelligence Bulls / Bears Ratio is now greater than 4:1 Bull / Bears;

2) American Association of Individual Investors Sentiment Survey last week had: Bulls - 27.1%, Neutrals - 46.1% and Bears 26.8%.

Ed Yardini tracks the 52 week moving average of Investor’s Intelligence Bulls / Bears Ratio which is now at 3.46 - the highest since records began in 1987! Vastly higher than in early 2000 or late 2007!! See chart in link below.



steve said...

william, that sentiment data is notoriously volatile. a 10% "correction" which is nothing but a blip would reverse the sentiment. all this talk about valuations, sentiment, momentum...I've heard ad nauseam for literally decades. make your investment decisions (preferably index funds) and leave them ALONE.

William said...
This comment has been removed by the author.
William said...

I'm not denying the wisdom of what you have said, Steve, including the use of index funds.

But at age 73 1/2, I am now 50% long stocks and 50% cash. No bonds. However, two years ago I was 124% long stocks.

Benjamin Cole said...

PPI in deflation. Retail sales stagnant. GDP Q1 dead in the water.
The Fed wants to raise rates?
I think if the Fed does raise rates, it will only lower long-term rates.

steve said...

william, cash is trash. suggest you look at HY bank loan funds, the sweet spot in bonds right now

William said...

Thanks for the suggestion, Steve.