The decline in long-term interest rates

Source: Council of Economic Advisers

Long term interest rates are trending down around the world. And it's not just since the great recession and financial crisis. The same trend has been going on for decades.

The Council of Economic Advisers just issued an excellent report surveying our understanding of this question. A blog post summary by Maury Obstfeld and Linda Tesar.

(Many other interesting CEA reports here. Occupational licensing is next on my in box.)

The report is really well done, for explaining the economic issues in clear simple terms, but without hesitating to use a model and an equation when necessary. If you're wondering how to keep your undergraduate or MBA class (heck, your PhD class) busy this week, this report will do the trick.

There is some grumbling in economics circles about the CEA and what role it should play, between Sunday morning talk show cheerleader for the Administration's policies vs. providing dispassionate  economic analysis to the Administration and country. This kind of report is the kind of CEA I cheer for.

I won't summarize the whole thing. Maury and Linda's blog post blog post does a great job of that, and you should just go read it. A few comments however.

1. Surprise surprise, the trend is a surprise. Hence, beware our current forecasts. This is not a criticism, it's just a fact. The best forecasts have been wrong in the past. They may well be wrong in the future.

2. Said: "The long-term interest rate is a central variable in the macroeconomy. A change in the long-term interest rate affects the value of accumulated savings, the cost of borrowing, the valuation of
investment projects, and the sustainability of fiscal deficits."

Unsaid: The surprise decline in long-term interest rates has been a boon to financing deficits. Current deficit forecasts use the current forecast of a return to higher interest rates. If this forecast is wrong once again, and real interest rates on government debt continue at rock-bottom levels, this will be a boon to "fiscal sustainability." Of course, the opposite is also true: If a trend nobody expected and everyone expects to reverse does reverse, then countries with big debts are in trouble.

3. The long term graph makes nicely a point that's been on the back of my mind lately. People typically assume that long term bonds should pay more then short term bonds, because they are riskier. But that's actually a puzzle: most bond investors hold their money for long periods of time, for which long term real bonds are less risky. It's hard, in fact, to get most term structure models to produce an upward-sloping yield curve.

It was not always so. In the 19th century, short term yields were consistently above long term yields.

The difference, of course, is inflation. In the 19th century we were on the gold standard, as noted in the graph. So long term bonds did not have inflation risk.  So, if inflation continues to die, or if our central banks go on a price level target, we might expect the same pattern to hold again. Which would be great for financial stability too. Short term debt causes runs and crises. If long term debt were cheaper, the inducement to finance short would be less.

4. Uncertainty. A message you read loudly between the lines is, that we have very good theoretical understanding of the various mechanisms that can move the trend in interest rates up or down, we (meaning "economic science") have really very little idea of the quantitative force of various mechanisms. By masterfully explaining each mechanism, and then patiently reviewing the vast literature that comes up with hugely different numbers for each mechanism, the point is made clearly, though between the lines.

They might go further. For example, the section on term premiums (the long rate is the average of expected future short rates plus a term premium) cites the latest studies and plots a line, but no standard error or other uncertainty band around that line. As this is an area I've written papers on, I know where the bodies are buried. Term premium estimates come down to forecasting regressions of future bond returns on current variables. Such regressions have huge bands of uncertainty. All forecasts and decompositions should have error bars. The only problem is artistic, as honest error bars would dwarf the forecasts. Well,  knowing what you don't know is real knowledge.

5. Forecasts. On p. 26, after this implicit devastating critique of the state of knowledge, "To illustrate our analyses, we illustrate different approaches to forecasting the long-term nominal interest rate, as is typically done twice a year in the CEA/OMB/Treasury Budget forecast and midsession review." A process for coming up with a number follows. Clearly, the message of the previous 25 pages is that conditioning decisions on a forecast, cranked out to two decimal places, is a bad idea. Economic policy should embrace uncertainty!

This is really a big deal. Much of the illusion of technocratic competence driving our regulatory state is reflected in absurdly accurate forecasts. The joke goes, we know economists have a sense of humor, because economists use decimal points. I'd love to see a Federal Forecast Accuracy Act: All forecasts made by every administrative agency shall include measures of forecast uncertainty. The CBO will evaluate all forecasts after the fact, and agencies shall be penalized when reality exceeds the stated uncertainty bounds more than half of the time.

6. The CEA ain't buying "secular stagnation," in its perpetual "lack of demand" interpretation.  (As a fact, it's undeniable. The question is the diagnosis and treatment.)  See p. 38.

7. In a report whose summary sections are  Fiscal, Monetary, and Foreign-Exchange Policies, Inflation Risk and the Term Premium, Private-sector Deleveraging, Lower Global Long-run Output and Productivity Growth, Shifting Demographics, The Global “Saving Glut”, Safe Asset Shortage, Secular Stagnation?, and Tail Risks and Fundamental Uncertainty, it is perhaps a bit petulant to complain of left-out factors but I will mention one.

The "supply side" part of the analysis is limited to productivity growth. Higher productivity growth leads to higher real interest rates in equilibrium, and (these days) vice versa. But it takes time and transition dynamics to accumulate capital.

One hypothesis that I learned from Larry Summers is that today's production function needs a lot less physical capital to produce the same productivity. A 1930s steel mill is a lot of accumulated savings. Facebook has nothing but a basketball court sized building full of 20-somethings coding while wearing headphones, and a really cool food court. The company is worth billions but it took comparatively little accumulated savings to start it up. If technology moves so that human, rather than physical capital is the heart of the K in F(K,L), productivity growth may determine interest rates in the long run, but there are lower interest rates on the transition path. Larry:

Ponder that the leading technological companies of this age—I think, for example, of Apple and Google— find themselves swimming in cash and facing the challenge of what to do with a very large cash hoard. Ponder the fact that WhatsApp has a greater market value than Sony, with next to no capital investment required to achieve it. Ponder the fact that it used to require tens of millions of dollars to start a significant new venture, and significant new ventures today are seeded with hundreds of thousands of dollars. All of this means reduced demand for investment, with consequences for equilibrium levels of interest rates.

(This is an update, thanks to email correspondent who found the quote.)

Update: Steve Williamson reminds us all that there is no "the" interest rate, and that the rate of return on capital is both stable and much higher than government bond yields. There is a risk premium, and it's big, and it varies over time. Practically all macro and growth theory forgets this fact. Since I've spent most of my career emphasizing the size and volatility of the risk premium, I should remember this reminder in every blog post. Thanks for pointing it out Steve!