Robert J. Shiller is the Arthur M. Okun Professor of Economics at Yale University, and Professor of Finance and Fellow at the International Center for Finance, Yale School of Management. In 2000, his book, “Irrational Exuberance,” predicted the bubble in technology stocks; in a subsequent edition and in a Wall Street Journal column in 2006, he warned of a bubble in real estate. He is the co-originator of the S&P Case Shiller Home Price Index, the leading index of home price values.
Dan Richards interviewed Shiller on January 4 over lunch at the Atlanta airport, returning from the annual meeting of the American Economic Association in Atlanta, GA. This interview is one of a series that Dan conducted at that conference. Dan is president of Toronto-based Strategic Imperatives.
You recently proposed that governments finance debt by issuing “trills.” A trill represents one-trillionth of the country’s GDP, and investors would receive a perpetual dividend of a trillionth of the country’s quarterly GDP. Can you describe how your proposal would work?
Right now people are concerned about how governments are going to finance their deficits. The US in particular has a huge financing need.
China borrows money now, even though in theory they are a surplus country. They do some borrowing just to keep the market working.
What are the advantages of your approach?
It would open up a whole new market to price discovery and risk management opportunities not known in history.
It would also enable investors to get a truly diversified portfolio, buying shares in countries instead of stocks. Foreign countries are wary of putting heavy investments in corporations based in other countries. If China put a huge investment into Canadian companies, then they would start to worry that the Canadian government’s policies toward those companies would waver, because it is the Chinese that are getting the profits. That’s why they put it in Treasury bonds.
Are there any other benefits to financing debt with trills?
It would open more cross-border investments, and it would lead to better risk management for governments, because their government debt is correlated not with what they owe but with GDP. If the GDP takes a downturn, their obligations go down, so it is risk management. If the advanced countries adopt this, it would set an example for developed countries.
Could a country like China issue trills?
China should eventually issue shares in its GDP. Today many people are wary of investing in Chinese companies – with trills they would be a little less wary China would get a huge price for its trills. They don’t know that, but there would be a market for it. They could get money for practically nothing.
Would inflation be a concern?
No. It’s like TIPS except more exciting. The yield on TIPS is now about 1%. Depending on the forecast for a country’s currency, the yield on trills could be even lower.
When a company goes bankrupt, there is a hierarchy for creditors within the capital structure. Where would trills fit within the capital structure of a country?
The bankruptcy of a country would be less likely for a developed country. Less developed countries won’t pay very much now if they issued trills because they are poor. They only have to pay a lot if they do well. If they don’t do well, they don’t pay a lot.
What are the barriers to implementation?
Some people say they don’t trust GDP numbers. They get revised, but people could probably live with the revisions. They tend to be relatively small.
There is some concern though that a government could deliberately fudge its GDP.
Would an oversight board be necessary?
The IMF already publishes GDP for every country. Today, we are better at measuring GDP. That’s one reason this hasn’t happened in the past – we didn’t have an identifiable “GDP” until we were well into the 20th century.
We are also better about surveillance. Right now, we have satellites recording the light from cities. They are continually orbiting. There was a recent NBER working paper that said light emission correlates pretty well to GDP. You could get a daily GDP estimate.
The US and Canada are not going to fudge their numbers. Neither has ever defaulted on its debt.
What have you done to advance this idea?
I went to the US Treasury and spoke to the Director of the Public Debt under George W. Bush, and I tried to get him interested. I just wrote to Larry Summers, who is someone I know. He started TIPS under [former Treasury Secretary Robert] Rubin.
What are the chances of getting the American Economic Association behind this idea?
I don’t know if the AEA has done anything like that as an organization, but it seems to me that you could get people to sign a letter.
Let’s talk about housing. What is the history of the Case Shiller index?
We have been working on it for over 20 years. We had a company start producing it in 1991, but we were producing it before that.
We sold the Case Shiller index to S&P, so we don’t have any commercial interest in it any more. We serve on an advisory board, out of our commitment to the quality of the data. They hope to license it for risk management, but they haven’t done that. So they are not making any money, but they are getting a lot of publicity.
We’re the only major index named after living people. Dow and Jones are dead. They don’t normally name indices after living people, but we had been producing it for 20 years before they bought it.
You have also attempted to start a futures market based on housing prices. What is the status of that effort?
It’s not very active. It’s one of those things like throwing a party. If everyone thinks everyone they know will be there, then they want to be there too. If not, then they won’t come. It became a self-fulfilling prophecy. It never got enough liquidity. I talked to potential investors who might have taken a look at it. The bid-ask spread was too wide. They called a dealer about placing a $10 million order. He said he couldn’t do that.
A lot of discussion centers on discrepancies between your numbers and those of the National Association of Retailers. Have you looked into that issue?
They have a strong seasonal bias. We think that is because big houses tend to sell in the summer. People in big houses tend to have children in schools, and they don’t want to move during the school year. You see this big uplift during the summer, but it’s not real.
They would also miss a number sometimes. We think that probably happens. They don’t announce why. It’s probably because there was a change in the mix of sales and they got some anomalous number and they don’t publish it.
At a macro level, how far are housing prices off?
About 30%, depending on the market.
What is your prognosis for real estate prices over the next year?
I don’t have a forecast. I don’t forecast. Since I produce the index, it’s not appropriate.
We are exceptionally uncertain now. We had a massive downtrend for three years. That peaked around April of last year. Was that a major turning point or some blip? I don’t have the answer to that. Our latest numbers, in October, did not show an increase. One theory is that it has to do with the stimulus – the house credit. Those are temporary things. Because of the shadow inventory that’s looming on the market and the direction of the economy, I am not bullish on housing.
To what extent is the US housing market buoyed by low interest rates?
It could be an issue but it’s not happening yet, although long rates are going up.
I’d like to talk about your cyclically adjusted price/earnings ratio. What is its history?
First of all, I don’t call this the cyclically adjusted price earnings ratio – I’m not sure where this description comes from. [see note at end of interview]
Around 1996, I began looking at stock valuations based on average earnings over the past ten years, as a result of work I was doing work for the Board of Governors under Alan Greenspan. John Campbell, a former student of mine who now is a professor at Harvard, and I presented it to Greenspan. We had a publication in the Journal of Finance around that time, where we used this P/E ratio. We showed that it predicts the stock market.
When you look at the P/E ratio, it tends to stay in a certain range. Something brings it back when it goes to the extremes – low or high. The question is, what brings it back? When the P/E is very high, it could be that the price (the numerator) goes down or the earnings go up. The efficient market theory would say it’s the latter, because the price doesn’t have any predictive value. The P/E ratio is predicting the returns.
We found that the opposite. When the P/E ratio is high, the price goes down.
We found that worked more strongly if you used the average earnings. One year is too volatile. In Graham and Dodd’s 1934 book Securities Analysis, they suggest using five- or even 10-year average. Some say it should be called the Graham and Dodd ratio. I don’t know if that’s fair, because it was just a throw-away idea for them.
Jeremy Siegel has written that in low inflation periods a higher price/earnings ratio is justified. Do you agree?
I’ve been through that with Jeremy, and I think he agrees it’s a small effect. I had a long discussion with him.
Another effect is the change in the payout ratio or percent of earnings that goes to dividends. As the payout ratio has decreased over the past thirty years, that’s left companies with more to invest and led to higher earnings. That’s also had a systematic effect on the P/E ratio.
Jeremy and I have been through this too. I don’t think that either of these issues is that big.
Based on your modeling today, is the market fairly valued?
I think it’s priced about right, maybe a little on the high side. The average is about 15, excluding the recent bubble.
If you look at past patterns, the stock market has been a reasonable investment, even when it’s at 20. You’ve got to put your money somewhere.
I’m still worried about the stock market, based on the economic crisis that we’ve had. I’m worried about a double-dip recession.
So are you in the double-dip camp for the US economy?
I sense not so much a sharp double dip, but disappointing performance. The unemployment rate will prove to be very hard to get down. The housing market may continue to decline because of the “shadow inventory.”
There may be a lot of defaults. I am worried that the default rate may go up. There is now a declining incentive to pay one’s mortgage.
About 25% of mortgages are underwater. People are paying out of their moral compunction. Those forces are weakening now. There is a new web site called www.walkaway.com, urging people to just walk away.
Do you have a view of prospects for US versus Europe, Japan or the emerging markets?
If you compare us to China, where I just was on my book tour, their stock market is so much more volatile than ours. I would predict more volatility in China.
What one piece of advice would you offer to President Obama in 2010?
The idea I would mention now are the trills, because we have a funding crisis. It’s time to talk about alternative funding.
The other thing I am pushing is the continuous workout mortgage. I like to see mortgages that are more advanced than in any country now, with a preplanned workout that is continuous, systematic and automatic. The current system penalizes people with high moral standards, who won’t default on their mortgages. They will continue to pay. It also creates crises. If they don’t get a workout, then their mortgage blows up. We need some kind of risk-managing mortgage.
Should the government discourage non-recourse mortgages?
I’d rather have a mortgage with an automatic workout. If home prices fall, your mortgage balance goes down. The mortgage issuers would like it, because they want to hedge their risk.
That’s why we created these futures markets. I had this book Macro Markets in 1993, where I had this vision for financial markets. You could trade things like home prices, GDP, medical costs, education costs, and other things that matter to people. That’s democratizing finance – trading things that matter to people.
If they do get traded, that could open a whole range of retail products. Most people won’t buy and sell GDP contracts. But other people would want the option of managing that kind of risk, and would buy and sell those kinds of products.
Note from Dan Richards: Two days after our conversation, I got an email from Robert Shiller indicating that he had Googled “Cyclically Adjusted Price Earnings Ratio” and found 250,000 citations. He said that, while he had not given his P/E ratio this title, the market had – and he planned to add this description to the spreadsheet on his blog.
Read more articles by Dan Richards
