Are stocks overvalued?

Well that should teach me never to go on vacation again.  I blogged pretty intensively for just over two years, from February 2009 through March 2011.  This frenzied activity was correlated with (or caused?) one of the great bull markets in American stock market history.  Then stocks started falling.  I had planned to return to blogging on July 5th, after a short vacation.  But with the market doing so poorly I felt duty bound to do something last week, and hence I came up with a few posts.  The effect on Wall Street was immediate and electrifying.

But (seriously) are stocks now overvalued?  Because I’m an efficient markets-type, the only answer I can give is no.   So why does Robert Shiller say yes?  Apparently because the P/E ratio is relatively high by historical standards.  And he showed that for much of American history investors did better buying stocks when P/Es were low than when P/E ratios were high.  Of course hindsight is 20-20.

I’d rather not get into the minutia of all the various ways of calculating P/E ratios.  And I have no idea where stocks are going from here.  Instead I’d like to focus on three arguments for relatively high P/E ratios in the 21st century American economy (however you’d like to measure them):

1.  Stocks have done very well since the 1920s, which suggests that 20th century P/E ratios were usually too low.

2.  American companies are making lots of money in the worst recession since the Great Depression.  This is partly because US multinationals are making huge profits in the developing world.  And this suggests that traditional market indicators based on the ratio of US corporate profits to US GDP may be outdated.  US GDP is no longer the relevant denominator.  So “E” may be relatively high for the foreseeable future.

3.  My most important argument is that low real interest rates might be the “new normal.”  The most striking characteristic of the US economy over the past decade is the unusually low level of both nominal and real interest rates.  And it’s not just because of the current “unpleasantness;” rates also fell to very low levels in the early 2000s.  Why have people missed this story?  I believe it’s because they’ve assumed the low rates are some sort of Fed policy, not a free market outcome.  But if the low rates since 2001 were an easy money policy, then why didn’t we see high rates of inflation and NGDP growth?  So money hasn’t been easy, which we should have obvious all along, given that INTEREST RATES ARE NOT THE PRICE OF MONEY, THEY ARE THE PRICE OF CREDIT.  And these low real interest rates should support a higher P/E ratio.

Why has credit been so cheap?  Perhaps it’s been special factors, but rates have been ultra-low in Japan since the early 1990s, and I don’t see any reason to assume that it can’t happen here as well.  (Of course Paul Krugman came up with this argument long before I did.)  All it would take is a combination of low demand for credit and high supply of credit.  For low demand, we have slowing population growth in the developed world, and fewer good investment opportunities (what Tyler Cowen calls fewer low hanging fruit.)  On the supply side, we have Asian countries that already have high saving rates at relatively low incomes, and which are rapidly becoming a larger share of world GDP.  Then think about the fact that the 1.3 billion people of China will soon live in a country with a falling population (i.e. similar demographics to Japan, but in a country 10 times larger.)

BTW, some people try to deny the “savings glut” argument by citing the relatively stable world savings ratio.  But if we are seeing both more global saving and less global investment, then you’d expect little change in the global savings/investment ratio.   (And yes, S does equal I at the global level.)

Once again, I’m not trying to predict stock prices, just suggesting why Shiller might be wrong.  Fans of Shiller might want to think about this.  He failed to give a strong buy call in March 2009, when stocks were at barely half current levels.  If he has the right model, why didn’t he scream out BUY in March 2009?

Yes, you can go back into history and find lots of other times when pundits like me claimed that high stock prices were the “new normal.”  But how many of those stock bubbles occurred in the midst of a severely depressed US economy, with very high unemployment?

How does it feel to be back blogging again?  Remember the look on Humphrey Bogart’s face as he was about to go back into the leech-infested waters for the second time?  I really wish the Fed would aim for a bit higher liquidity levels, so that we can float free from these reeds, er, I mean a bit faster growth in AD, so I can have my life back.

I’ll do a couple more posts this evening, and then go full blast starting tomorrow.

Update:  I guess I’m still rusty.  Commenter Steven pointed out that more aggregate earnings (point #2) doesn’t raise the P/E ratio.  When I wrote the post I was actually trying to explain the high ratio of stock values to GDP, but ended up using the P/E ratio terminology.  However Shiller uses a 10 year moving average P/E ratio, so his method arguably overlooks the importance of the recent growth in the foreign earnings of US multinationals.



21 Responses to “Are stocks overvalued?”

  1. Gravatar of Steven Steven
    5. July 2011 at 18:50

    E isn’t GDP, it’s earnings (i.e., net income of the company). So if the effects of globalization show up in current earnings, there’s no reason to expect it to change P/E ratios. So you need to argue that globalization increases future earnings more than it increases current earnings (i.e., that globalization increases the value of the company by more than it increases current earnings). That’s plausible, but sufficiently non-obvious that you really need to develop a more complete argument.

  2. Gravatar of Nick Rowe Nick Rowe
    5. July 2011 at 18:56

    It’s great to have you back, Scott!

    So, that natural rate of interest has fallen, so equilibrium real asset prices should be higher. Stocks, farmland, oil, gold, and, errrr, houses!

    And asset prices were going towards their new equilibrium values in 2005/6/7, then (take your pick): the markets chickened out; the Fed screwed up by lowering expected NGDP. So asset prices fell. Now some of those asset prices are trying to climb up towards that equilibrium again.

    (Andy Harless had a very good related post about a year or so back.)

  3. Gravatar of Scott Sumner Scott Sumner
    5. July 2011 at 19:38

    Steven, Thanks for pointing that out. I didn’t present my argument effectively. I was actually (in point two) trying to explain high stock prices, not high P/E ratios. I suppose I was also thinking (implicitly) along the lines that you are suggesting–Shiller uses a 10 year moving average, and the huge rise in US multinational earnings in places like Asia is a fairly recent phenomenon.

    I’ll add a correction.

    Nick, Yes, I am claiming that all sorts of asset prices should be higher, and in most countries real estate and commodities (like gold) are higher. In addition, stock prices are pretty high in developing countries, when you consider the severity of the recession.

    Andy is someone who should do more blogging.

  4. Gravatar of Richard A. Richard A.
    5. July 2011 at 20:52

    Here is a FRED graph of the S&P 500 divided by nominal GDP–

  5. Gravatar of Tom Powers Tom Powers
    5. July 2011 at 21:06

    Re: Shiller, the fact that P/E ratios forecast returns can fit within the efficient markets framework, since the P/E ratio might just be an indicator of people’s (rational) discount rates. When those rates are low, we get low returns.

    In fact, I’d be surprised if discount rates didn’t change across time, given demographic shifts and shifting tastes and risk aversion levels. It’s also not too surprising that some metric (P/E) could indicate the underlying level of discounting.

  6. Gravatar of Morgan Warstler Morgan Warstler
    5. July 2011 at 23:18

    There are giant bushels of low hanging fruit:

    1. 12M homes to be sold in $1 auctions.
    2. Closing the Post Office forces 12% of the economy to modernize (advertise elsewhere) and old people to learn to use the internet to survive. And apparently it is something called “green,” and at the same time saves newspaper publishers for quite some time.
    3. $500B in wage savings from gutting public employees and automating government.

    The reality is “growth” comes from getting rid of buggy whips and postal employees / public employee unions are both Trillion dollar buggy whips.

    Folks who don’t see the fruit, just really like buggy whips.


    Doesn’t Shiller need to say stocks are overvalued, so people will start buying investment houses? Isn’t he just selling his book?

  7. Gravatar of Erik Erik
    6. July 2011 at 01:41

    If real interest rates are low should not the return on equities be lower as well? If returns on equities are a premium over long term interest rates of 3-5%. This fits well will the idea that we have an oversupply of saviings. This does not say that equities are overvalued but might well mean that the returns to both stocks and bonds might be very low going forward? And since volatility hardly seems to be lower investors might be in for a wild ride without much return to show for the risks they take.

  8. Gravatar of Brendan Brendan
    6. July 2011 at 06:15

    Tom, you said:

    “Re: Shiller, the fact that P/E ratios forecast returns can fit within the efficient markets framework, since the P/E ratio might just be an indicator of people’s (rational) discount rates. When those rates are low, we get low returns.”

    The problem is that investors’ return expectations are consistently highest after prolonged bull markets when risk premiums and subsequent returns are actually lowest- and vice versa. That is pretty devastating to the rational time-varying risk premium story. (Financial economists actually make the same mistake on average too.)

    Scott thinks markets are more efficient than they really are, but he is absolutely right (as usual) about all three points he made; all important, true and overlooked. The recovery in profit margins to levels well above historical norms has shocked lots of bearish investors who thought the high margins of 2005-2007 were a fluke driven by commodity and housing bubbles. And lots of these guys fail to even address the impact of much lower real rates on P/E’s.

    My biggest question is whether investors yet realize that current valuations imply no better than 5-6% long-term nominal returns. In a taxable account that implies 1-2% real returns. At what point do zero real risk free rates and 1-2% real returns on stocks ignite the housing market?

  9. Gravatar of D. Watson D. Watson
    6. July 2011 at 06:57

    Glad to have you back.

  10. Gravatar of Benjamin Cole Benjamin Cole
    6. July 2011 at 08:31

    This is superior blogging. Sumner, this is great.

    Some people are calling for higher interest rates, as if artificially inducing a fever will cure anemia.

    Please, policy-makers of the world: Read Scott Sumner.

  11. Gravatar of Jeff Jeff
    6. July 2011 at 08:34

    All of us leeches are glad you’re back.

  12. Gravatar of ssumner ssumner
    6. July 2011 at 11:04

    Richard A. Thanks, but that’s not quite the right ratio. You’d need total value of US stocks divided by GDP. (Unless S&P 500 is a stable fraction of total stock market cap.–in which case it’s fine.)

    Tom, Good point. His argument is that the shifts are too large to be plausible. But obviously I don’t agree with his conclusion.

    Morgan, I think Shiller claims houses are also overvalued.

    You said;

    “Folks who don’t see the fruit, just really like buggy whips.”

    I was tempted to make a joke, but this is a family blog.

    Erik, You points are valid, but it’s important to distinguish between levels and rates of change. Let’s say the public gradually comes to accept higher P/E ratios. During the transition period stocks will rise sharply. I see 1982-2000, or 2007, as that sort of transition. From that point forward rates of return will be low, as you must pay a high price for any given dividend flow.

    I’m saying stock prices and P/Es will be high, not rates of return.

    Brendan, Good points. But also consider that the same calculations that you apply to real, after-tax stock returns imply extremely low (often negative) real after-tax T-bond returns. And real estate has lots of other issues, and is therefore not as close a substitute to stocks. Still your points are well taken–I am working on a new post linking the immigration crackdown and housing slump.

    Thanks D. Watson

    Thanks Benjamin

    Jeff, I was wondering if anyone would notice that implication. Of course I was only thinking about those annoying MMT commenters who occasionally come over here to insult me. 🙂

  13. Gravatar of Dave Dave
    6. July 2011 at 14:10

    “…then why didn’t we see high rates of inflation…?”

    Because the official inflation measures didn’t include bubbling home prices…

  14. Gravatar of Doc Merlin Doc Merlin
    6. July 2011 at 16:19

    “Why has credit been so cheap? ”
    Because the fed forced rates lower and lower! Its to the point where no one is buying treasuries anymore except the fed.

    Seriously, the credit market is in a really big slump because the fed forced down rates so low that no one wants to invest, and debt levels are so high (we didn’t get the sudden delevering we needed) that people can’t borrow money safely either.

    “I really wish the Fed would aim for a bit higher liquidity levels, so that we can float free from these reeds, er, I mean a bit faster growth in AD, so I can have my life back. ”

    NO NO NO! Look, there is all the liquidity in the market anyone could want, seriously, we are drowning in liquidity. The problem isn’t an illiquid market, but that rates are being held down too low because of public finance reasons. (if they were allowed to rise we would have other problems, but thats a different topic)

  15. Gravatar of David Stinson David Stinson
    6. July 2011 at 17:44

    Hi Scott.

    NIce to have you back but of course it just means I have to spend all kinds of time reading your posts. (It may be difficult to discern but there’s actually a compliment in there).

    I’m probably missing something obvious here but I don’t understand this paragraph:

    “BTW, some people try to deny the “savings glut” argument by citing the relatively stable world savings ratio. But if we are seeing both more global saving and less global investment, then you’d expect little change in the global savings/investment ratio. (And yes, S does equal I at the global level.)”

    In the absence of monetary disequilibrium, how could one be seeing both more global saving and less global investment?

    You also noted that rates have been ultra-low in Japan since the early 1990s, which I assume was the genesis of the carry trade. What is your take on why the carry trade was/is sustainable, if not some sort of continued monetary disequilibrium?

  16. Gravatar of RN RN
    6. July 2011 at 19:25

    Shiller called the internet bubble. He the called the housing bubble well ahead of others.

    You’ve done…what, exactly?

  17. Gravatar of Thursday 7atSeven: debt ceiling debate | Abnormal Returns Thursday 7atSeven: debt ceiling debate | Abnormal Returns
    7. July 2011 at 03:04

    […] Are stocks overvalued, ala Shiller?  (The Money Illustion) […]

  18. Gravatar of Alban Alban
    7. July 2011 at 04:23

    First time reader. Thanks for the blog; a question: can you explain money velocity?

  19. Gravatar of Scott Sumner Scott Sumner
    7. July 2011 at 08:32

    Dave, That doesn’t help at all, as housing prices are back down (in relative terms) to 2001 levels. So with or without housing there has been little total inflation during 2001-11, and modest NGDP growth. And all this despite low real rates.

    Doc Merlin, No, markets set rates, not the Fed. Consider rates on longer term debt, also quite low. Yet easy money tends to raise long term rates.

    David, That was poorly worded on my part. I meant that the savings schedule shifted right, and the investment schedule shifted left. The price fell, but quantities were little changed.

    RN, You said;

    “Shiller called the internet bubble. He the called the housing bubble well ahead of others.

    You’ve done…what, exactly?”

    Totally wrong, he did not call the internet bubble. He said sell stocks in 1996, when the market wasn’t overvalued. He was wrong. Sure, if he keeps saying sell over and over and over, eventually he will be “right.”

    I don’t claim to be a Nostradomus, so it’s irrelevant what I’ve called. I’m not superstitious like most of my fellow economists, I don’t think we can beat the markets.

    Alban, Sure, velocity is the ratio of NGDP to the money supply, however defined. There is a different velocity for each definition of money.

  20. Gravatar of 123 123
    7. July 2011 at 10:14

    Tobin’s Q ratio tracks Shiller’s P/E10 ratio very closely. Do your arguments for high P/E also apply to high market value of stocks vs. their replacement value?

  21. Gravatar of ssumner ssumner
    8. July 2011 at 17:39

    123, I’m not sure, because Tobin q is hard to do right. There’s lots of intangible intellectual capital in modern corporations. I’d guess there’s some correlation, so I suppose I do think Tobin’s q will be a bit higher than average.

Leave a Reply