The low interest rate century

During the Great Recession, I argued that low interest rates were the new normal and that during the 21st century people would be constantly complaining about “bubbles”. But even I never envisioned rates being this low.

Ten-year bond yields are still 20 basis points above the 2016 lows, while 30-year bond yields have been hitting all-time record lows. Why the difference?

In my view, the 10-year yield is a bit more influenced by cyclical factors, and the economy was a bit weaker in 2016 than it is today. In contrast, the sharp fall in the 30-year bond yield reflects the market gradually realizing that low rates are not just a passing fad, but rather are the new normal.

I thought we’d cycle between 0% and 3% over the business cycle, now it looks more like a 0% to 2% cycle might be the new normal. This cannot be explained by inflation, which isn’t much different from what it was in the late 1990s.



16 Responses to “The low interest rate century”

  1. Gravatar of Cameron Blank Cameron Blank
    11. October 2019 at 10:02

    As a saver who puts a lot in broad index funds I’m curious whether you expect long-term stock market returns to be significantly lower in the foreseeable future? I doubt we’ll continue to see 7% but is a 4% return too much to expect?

  2. Gravatar of John Hall John Hall
    11. October 2019 at 10:57

    It definitely is interesting that long-run interest rates are so much lower than expected NGDP growth.

  3. Gravatar of Tuharsky Tuharsky
    11. October 2019 at 11:14

    Given that you think that, at least in the short run, shocks to NGDP translate into shocks to RGDP, and if one expects RGDP to be correlated to stock market returns, doesn’t it make sense for long term interest rates to be “too low” because you are not just buying a bond, but also a put (or, if you prefer to think about it in terms of NGDP, an American-style call on the currency?) In other words, one can cancel out undiversifiable stock market risk by buying bonds, so bond prices should have a real return below the long-term natural rate to reduce the expected return of a diversified bond/stock portfolio to that equilibrium rate. Moreover, the higher expected volatility of NGDP, the more long-term bonds should trade at a premium, since the more one would be willing to pay for protection from downward NGDP shocks.

    Given that (1) market actors may have learned since 2008 to expect higher NGDP volatility and (2) global central bank policy seems to be making long-term bonds an ever more effective form of such protection, is it not surprising that we see ever lower yields on government bonds?

  4. Gravatar of ssumner ssumner
    11. October 2019 at 11:15

    Cameron, That makes sense, but of course I have no claim to expertise on predicting stock market returns. There are actually two issues here:

    1. The long run steady state return.
    2. Will there be an additional one-time adjustment upward in the P/E ratio due to lower rates?

    I can’t answer that. And what happens if Warren wins in 2020?

    John, Yes, that’s really what has surprised me. NGDP growth has fallen, but rates have fallen even more.

  5. Gravatar of ssumner ssumner
    11. October 2019 at 11:16

    Tuharsky, That makes sense, but do investors actually expect more NGDP volatility going forward? I actually expect less than pre-2008.

  6. Gravatar of Matthew Waters Matthew Waters
    11. October 2019 at 11:18

    The issue is that “safe” countries have declined as a percentage of worldwide GDP. As GDP increases in China, India, Latin America, even Africa, the savings demand from those countries goes to developed countries.

    Over the course of decades, low interest rates could change if emerging markets get more trust from savers in their own countries.

  7. Gravatar of Benjamin Cole Benjamin Cole
    11. October 2019 at 16:19

    Think globalized capital markets, and large surpluses of global capital. 30-year Treasury rates are not set by only what happens in the United States, 30-year Treasury rates are a global rate.

    This is why Bank of England Governor Mark Carney and many others say that central banks are armed with popguns when it comes to setting interest rates. If you have globalized capital markets, then long-term interest rates are set globally.

    Really, if German sovereigns are paying negative interest rates, why would you wonder if US long term rates are also falling?

  8. Gravatar of Benjamin Cole Benjamin Cole
    11. October 2019 at 18:58

    Tinfoil Hat Time:

    Okay, so I am puttering around the house, aluminum fedora on, listening for the latest signals from Mars, when I read that the US Federal Reserve will increase its balance sheet by $60 billion monthly, but this not quantitative easing.

    So, on top of the $200 billion in Treasuries the Fed bought recently, add $60 billion monthly through mid-way 2020.

    But this is not QE, as the Fed is buying short-term bills. Not QE!

    Okay, money is a fungible commodity, and we have globalized capital markets.

    Still, the experts are saying this latest round of non-QE will not have an effect, macroeconomically speaking, as It is not monetary policy. No!

    I guess the experts say the Fed can more or less cash-out short-term national debt at any time, and we still not get serious inflation, in any, or any stimulus either. A nothingburger. I guess Japan is the poster-boy for that.

    You know, a few hundred billion dollars here, and few hundred billion dollars there….just technical fudging.

    So…what does this say about the effectiveness of Fed policy in terms of real macroeconomic output?

  9. Gravatar of Mike Sandifer Mike Sandifer
    12. October 2019 at 06:34


    Do you think there’s a theoretical limit to how low real and nominal rates can go in a growing economy? How about in a contracting economy?

    And then a follow-up, can rates go low enough that you begin to seriously question the market monetarist claim that rates alone aren’t indicative of the stance of monetary policy, or perhaps rates versus NGDP growth?

    As you know, I don’t buy the market monetarist claim that nominal real rates are so ambiguous. I suspect there are limits to how rates behave, beyond which it starts to become difficult to deny that they alone aren’t indicating something about the monetary policy stance.

    I’m not convinced I’m correct either though.

  10. Gravatar of Mike Sandifer Mike Sandifer
    12. October 2019 at 06:51

    The above should read “they alone aren’t indicating something about the monetary policy stance”.

  11. Gravatar of Mike Sandifer Mike Sandifer
    12. October 2019 at 06:53

    “they alone are indicating something about the monetary policy stance.”

    Autocorrect is killing me.

  12. Gravatar of ssumner ssumner
    12. October 2019 at 11:02

    Mike, No, I don’t think there is a theoretical limit on interest rates.

  13. Gravatar of Brian Donohue Brian Donohue
    13. October 2019 at 07:28

    Scott, it seems to me that nominal rates start facing resistance at 0% in the presence of things like currency (or other strategies for preserving purchasing power), so I’d be surprised if nominal rates could go significantly negative.

  14. Gravatar of ssumner ssumner
    13. October 2019 at 11:24

    Brian, I agree, I was referring to real rates.

  15. Gravatar of Mike Sandifer Mike Sandifer
    13. October 2019 at 16:02


    The question I should have asked is, what, if any, limits are there to how low the neutral interest rate can go?

  16. Gravatar of dtoh dtoh
    14. October 2019 at 15:29

    So here is a hypothesis. Maybe inflation doesn’t fully capture quality gains, and therefore we are mis-measuring real returns.

    Simple thought experiment. Society produces bananas. At the margin, we’ll give up 10 bananas of consumption in order to increase our annual production of bananas by 1 banana. On the other hand though, we’ll give up 100 bananas of consumption in order to produce 1 more extra-super-delicious banana.

Leave a Reply