Sunday, November 23, 2014

The Economic Costs of Fear

BERKELEY – The S&P stock index now yields a 7% real (inflation-adjusted) return. By contrast, the annual real interest rate on the five-year United States Treasury Inflation-Protected Security (TIPS) is -1.02%. Yes, there is a “minus” sign in front of that: if you buy the five-year TIPS, each year over the next five years the US Treasury will pay you in interest the past year’s consumer inflation rate minus 1.02%. Even the annual real interest rate on the 30-year TIPS is only 0.63% – and you run a large risk that its value will decline at some point over the next generation, implying a big loss if you need to sell it before maturity.

So, imagine that you invest $10,000 in the S&P index. This year, your share of the profits made by those companies will be $700. Now, imagine that, of that total, the companies pay out $250 in dividends (which you reinvest to buy more stock) and retain $450 in earnings to reinvest in their businesses. If the companies’ managers do their job, that reinvestment will boost the value of your shares to $10,450. Add to that the $250 of newly-bought shares, and next year the portfolio will be worth $10,700 – more if stock-market valuations rise, and less if they fall.

In fact, over any past period long enough for waves of optimism and pessimism to cancel each other out, the average earnings yield on the S&P index has been a good guide to the return on the portfolio. So, if you invest $10,000 in the S&P for the next five years, you can reasonably expect (with enormous upside and downside risks) to make about 7% per year, leaving you with a compounded profit in inflation-adjusted dollars of $4,191. If you invest $10,000 in the five-year TIPS, you can confidently expect a five-year loss of $510.

That is an extraordinary gap in the returns that you can reasonably expect. It naturally raises the question: why aren’t people moving their money from TIPS (and US Treasury bonds and other safe assets) to stocks (and other relatively risky assets)?

People have different reasons. And many people’s thinking is not terribly coherent. But there appear to be two main explanations.

First, many people are uncertain that current conditions will continue. Most economists forecast the world a year from now to look a lot like the world today, with unemployment and profit margins about the same, wages and prices on average about 1.5% higher, total production up roughly 2%, and risks on both the upside and the downside. But many investors see a substantial chance of 2008 and 2009 redux, whether triggered by a full-fledged euro crisis or by some black swan that we do not yet see, and fear that, unlike in 2008 and 2009, governments would lack the power and will to cushion the economic impact.

These investors do not view the 7% annual return on stocks as an average expectation, with downside risks counterbalanced by upside opportunities. Rather, they see a good-scenario outcome that only the foolhardy would trust.

Second, many people do see the 7% return on stocks as a reasonable expectation, and would jump at the chance to grab it – plus the opportunity of surprises on the upside – but they do not think that they can afford to run the downside risks. Indeed, the world seems a much more risky place than it seemed five or ten years ago. The burden of existing debts is high, and investors’ key goal is loss-avoidance, not profit-seeking.

Both reasons reflect a massive failure of our economic institutions. The first reason betrays a lack of trust that governments can and will do the job that they learned how to do in the Great Depression: keep the flow of spending stable so that big depressions with long-lasting, double-digit unemployment do not recur. The second reveals the financial industry’s failure adequately to mobilize society’s risk-bearing capacity for the service of enterprise.

As individuals, we appear to view a gamble that has a roughly 50% chance of doubling our wealth and a roughly 50% chance of halving it as worthy of consideration – not a no-brainer, but not out of the question, either. Well-functioning financial markets would mobilize that risk-bearing capacity and put it to use for the benefit of all, so that people who did not think that they could run the risks of stock ownership could lay that risk off onto others for a reasonable fee.

As an economist, I find this state of affairs frustrating. We know, or at least we ought to know, how to build political institutions that accept the mission of macroeconomic stabilization, and how to build financial institutions to mobilize risk-bearing capacity and spread risk. Yet, to a remarkable degree, we have failed to do so.

  • Contact us to secure rights


  • Hide Comments Hide Comments Read Comments (11)

    Please login or register to post a comment

    1. CommentedGreg Rushing

      "Both reasons reflect a massive failure of our economic institutions. The first reason betrays a lack of trust that governments can and will do the job that they learned how to do in the Great Depression: keep the flow of spending stable so that big depressions with long-lasting, double-digit unemployment do not recur. The second reveals the financial industry’s failure adequately to mobilize society’s risk-bearing capacity for the service of enterprise."

      Yes. That's the problem. Investors are shunning stocks because they are essentially upset that current government policies aren't closer to FDR's New Deal policies.

    2. CommentedJason Cawley

      Yes we know how to build such institutuons, the technical capacity is not a problem. But those running our institutions do not want to use them for those purposes.

      Our macroeconomic policies are run to ensure re-election and reward supporters, and our financial institutions have in the recent past been run, to too large an extent, to extract payments to their operators from their clients.

      Added to the disfunction of each is the greater disfunction from their interaction after each fails - the political system turns to using the banks as a punching bag for blame, and via defaults into dumping ground for socialized losses.

      Everyone knows that the only thing that has ever made capital cheaper is paying on the nail as contracted, yet entire regions use the public credit simply to extract one time transfers from creditors, with no intention of actually repaying. Then regulators wonder why capital flees, when it doesn't evaporate outright.

      You are right to be frustrated, but wrong to imagine that any of it was ever a problem of knowedge. It is a problem of character and of politics, and a very old one.

      Why do so many major banks trade at half or less of book? Hint, check out the bulleyes on their backs...

    3. CommentedDavid Doney

      Regarding the right return rate to use for the S&P 500, I'd be curious what time period and assumptions he is using.

      From January 1990 (S&P at 339) to May 1 2012 (S&P at 1313) we had about a 6.2% nominal annual return, excluding dividends (this is the monthly CAGR x 12). So assuming dividends and inflation offset, he's in the ballpark but a little high.

      However, if you look at the chart below, the S&P 500 is still below its 2000 level. It approached 1,500 in January 2000, then fell to 815. It made it to 1,526 in July 2007, then fell to 797. It got to 1408 recently,
      then fell back again a bit.

      In other words, its been in a trading range with two nearly 50% declines for the past decade! That is a huge amount of risk!

      What does this suggest? Buy stocks with sustainable 4-6% dividend yields (utilities and oil) and hold them. Not much point in trying for growth.^gspc;range=19900101,20120530;compare=;indicator=volume;charttype=area;crosshair=on;ohlcvalues=0;logscale=off;source=undefined;

        Portrait of J. Bradford DeLong

        CommentedJ. Bradford DeLong

        Yes, there is a lot of risk involved in investing in stocks. But there is also an even much larger possibility of upside gains...

    4. CommentedRichard Foosion

      BTW, didn't Fight the Fed Model by one Clifford Asness include that "traditional p/e is what matters in forecasting long-term stock returns."

        CommentedClifford Asness

        Your comments about 1-year P/Es were reasonable, even if I disagree. I was probably too harsh in my initial comment, I admit. But DeLong doesn't even say "I'm using a figure on the high bullish side of the current debate", he simply presents his method, a flawed method in my eyes (it's amazing how fast the bulls drop 1-year P/Es when E is depressed), as fact and moves on.

        CommentedRichard Foosion

        That's what I get for reading too quickly. At least you didn't push back on my other comments.

    5. CommentedRichard Foosion

      The earnings yield based on one year earnings is around 7%. Shiller's PE10 might be better methodology (and Brad has used it in the past), but regular p/e isn't crazy.

      What we really want to know is the price of future earnings, so we're left with using some imperfect proxies, such as e/p or dividend yield plus growth, which likely have an error well in excess of the extra 2%

    6. CommentedClifford Asness

      The repeat of the argument at the end was a bad cut and paste. The shot at liberals was just mean. Go figure. But DeLong really can't add.

    7. CommentedClifford Asness

      The S&P 500 yields 7% real? How do you get that? You give an example that is just a tautology - returns are 7% real because you give numbers that sum up to 7% real.

      The Shiller P/E is about 21. Most who've looked at this would put the real E[r] of the S&P 500 at about 5% real accordingly. That is not a dangerous level, but it's low versus history (the Shiller P/E is about 60th percentile expensive (high) since 1960 so the earnings yield is low versus history, a bit worse if you look even further back in time).

      Alternatively dividends yields are about 2.3%. If you add in historic real growth of about 1.5% you get almost (rounding up) 4%. If you'd like to add another 1% for super-optimism (like extra growth or buy backs or something else that's never worked) fine, 5%, again that's optimistic but I'll give it to you.

      These are now fairly standard forecasts among academics who look at the ERP. How on Earth do you get an extra 2% (a huge number!) to get to 7%, and how do you do that w/o explanation?

      Your central point, if stocks vs. bonds is your central point, is not wrong, just wildly overdone w/o explanation. While as described above, equities are around historical mean valuation (again, a bit expensive), bonds are indeed very expensive. But looking at the difference between stocks and the 10-year treasury, but NOT using your crazy 7% real for stocks, the differential is indeed pro-stocks, but at about 75th percentile since 1960. That is not an extreme reading and not consistent with the histrionics in your artcile.

      Do better. Explain more.

      Now, for stocks vs. bonds he's at least directionally right, but w/o giving numbers, he's way off in his rhetoric. Comparing the E[r] through the Shiller E/P to the agreed very low real yield on bonds the difference is in the 75th percentile back to 1960. High, but hardly worth a histrionic article.

      Just thought I'd wax quantitative for a moment.