Thursday, August 23, 2012

Further Evidence on the Fed's Superpower Status

I have made the case many times that the Federal Reserve is a monetary superpower. The Fed has this power because it manages the world's main reserve currency and many emerging markets are formally or informally pegged to dollar. As a result, its monetary policy gets exported to much of the emerging world. The ECB and Bank of Japan are also influenced by the Fed's decisions because they are careful not to let their currencies becomes too expensive relative to these dollar-pegged currencies and the dollar itself.  U.S. monetary policy, consequently, gets exported to the Eurozone and Japan as well.  

This understanding helps explains why there was a global liquidity glut during housing boom period and suggests that some of the "global saving glut" was simply a recycling of loose U.S. monetary policy.  It also implies that the Fed could now do a lot of good for both the U.S. and Eurozone economies if were to adopt something like a NGDP level target.  Based on this view, the global economy sorely needs the Fed to wake up from its slumber. Fed Chairman Ben Bernanke admitted as much in one of his classroom lectures earlier this year.1

Chris Crowe and I developed this monetary superpower hypothesis in a paper that is now in my newly published book. In a new paper, Colin Gray has gone even further by formally motivating this hypothesis in a rational expectations model.  He also has provided more robust empirical evidence for it.  Here is his abstract:
Between 2002 and 2006, the United States Federal Reserve set interest rates significantly below the rates suggested by well-known monetary policy rules. There is a growing body of research suggesting that this helped fuel an excess of liquidity in the U.S. that contributed to the 2008 worldwide financial crash. What is less well known is that a number of other central banks also lowered interest rates during this period. An important question, then, is what role the Federal Reserve played in influencing other central banks to alter their own monetary policies, which could have magnified the Fed’s actions in creating global liquidity. This paper addresses the issue by showing how spillovers in central bank behavior occur in theoretical rational expectations models. It then establishes empirically how U.S. monetary policy actions affect the actions of other major central banks, particularly in terms of interest rates and currency interventions. The data suggest that the U.S. lowering its policy rate, in general or in reference to a specific monetary policy rule, influences other central banks to lower their own policy rates and intervene in currency markets, even when controlling for worldwide macroeconomic trends. Finally, this paper shows that spillovers from U.S. actions are partially responsible for the worldwide lowering of interest rates and the increase in currency reserves in the early 2000’s that may have contributed to the subsequent worldwide liquidity boom.

The findings of this paper deserve further discussion.  In particular, how should the Fed operate given it has this monetary superpower status?  Is there a way to do U.S. monetary policy that maximizes macroeconomic stability for both the United States and the rest of the world?  These are important questions that will become more important over time as the global economy continues to integrate. 

1Okay, he really did not admit it, but it was implied in his acknowledgement that Chinese monetary policy is influenced by U.S. monetary policy.


  1. Isn't there a case that some effort should be made to break US monetary super power status? After all, the world is not an OCA. I think both a carrot and stick might be able to loosen the dollar peg zone, greater guarantees of monetary liquidity swaps and IMF/WTO action for offenders.

  2. OGT, I agree the world is not an OCA. But to end the Fed's global influence would require a lot of change by countries linked to the dollar. I would love to hear more about how your proposals would do that.

  3. David- To the extent that countries are self insuring in the after math of the Asian crisis I think central bank swap lines and IMF reform could influence them to reduce reliance on that strategy.

    For those, like China and the oil states that are using the dollar peg for mercantile purposes, I think the IMF should refer them to the WTO as trade violations. Even lost cases would likely have a chilling effect.

    Lastly, the US should have a weaker dollar right now as the best way to pursue a NGDP target. The Fed and Treasury should use the expectations channel by setting aside any strong dollar talk, and make it clear Fed policy is only concerned with its legal mandate not the asset values of foreign dollar holders.

    Granted there's risk in the whole strategy and some transition management, but institutionalizing the present situation seems like a recipe for repeated crisis.

  4. This is a great post; it was very informative. I look forward in reading more of your work.Also,I made sure to bookmark your website so I can come back later.I enjoyed every moment of reading it.

    shot for slim

  5. Interesting post.

    I have to say, I don't think the Fed was too easy leading into the housing boom. I like prosperity. Inflation was low. Growth was okay, could have been better. The Fed up until 2008 was okay.

    The Fed was too tight after the bust (or it caused the bust by tightening, fearful of commodities inflation).

    Oddly, commodities inflation cannot be sustained: people start using less oil, or making more of it.

    Gold may be an exception, and that is a reason to ignore gold.

    I think the USA economy has been inoculated against inflation since the 1970s, by global trade and a deunionized private-sector labor force, and rapid technical innovations.

    Name an industry that has pricing leverage, or name a single private-sector union asking for "more." (Samuel Gompers is way gone).

    Congrats on your book!

  6. This comment has been removed by the author.

  7. spillovers are irrational, thus how could they be "rational"

  8. I find the argument that the Fed had too loose policy during the housing boom very odd, because that is basically arguing that our current situation of high unemployment would have been better than having a housing bubble. For policy not to be "too loose" during 2001-06, then interest rates would have to have been higher and thus unemployment would have been higher.

    In fact, the 1% interest rates were still too tight according to NGDP. While we technically did not have two straight quarters of GDP decline, NGDP did not keep with trend for years and the anemic job growth 2001-04 bears that out.

    So what then? Should interest rates have been zero? Negative through IOER penalty, or with significant QE? The immediate thought is probably such policy would have made the housing bubble even deeper.

    But here's another way of looking at the composition of NGDP. If we take out government spending and net exports, we have the following equation:

    Private Demand = Consumption + Investment

    During the "Global Savings Glut," there was clearly significant irresponsible investment, where investors lent $1 to get 50 cents back. In my mind, this global savings glut had two main factors:

    1. What you mentioned, the flood of savings from Asia and Petro-centric countries. These countries had much higher savings demand than westerners typically have.

    2. Increasing income inequality, in the US and elsewhere. Bill Gates might fine lending $1 to get a guaranteed $1.01 back a year from now. A homeless man, however, would need 1,000% yearly interest or more to not see the money for a year.

    These are the big reasons real interest rates have gone down a good bit since the 80's. Bill Gates is fine getting 1% a year back even if it doesn't keep up with inflation. I'm agnostic to income inequality, in and of itself, being an issue, but inequality does shift the savings demand curve to the right. With interest rate on the y axis and total savings/investment on the x axis, greater inequality shifts the upward-sloping demand curve to the right. If the supply curve is the same, real interest rates go down and total savings/investment go up.

    Alright, so that explains why interest rates have been so damn low. What about the housing bubble? The answer is that the above interest rate model says that total savings/investment will go up with higher savings demand and lower interest rates. However, it does not say WHERE the savings will go.

    Due to behavioral factors causing bubbles, poor banking and finance regulation, AAA rating giving perceived risk-free assets, whatever, a lot of that savings/investment have gone to irresponsible investment. To go back to the equation above, let me split private investment into two factors:

    Private Demand = Consumption + Positive NPV Investment + Negative NPV Investment

    By "Positive NPV Investment," I mean that the Efficient Market Hypothesis applies and the price of the asset "reflects all available information." That means some of these investments may turn out to have Negative NPV, but they balance out to a risk-adjusted return on investment with the expected stochastic factors.

    The "Negative NPV Investments" are investments which have a rate of return below their correct risk-adjusted rate. The primary stock market for Internet IPO's and the original subprime MBS fell into this category. Meanwhile, secondary market trading of Internet stocks and derivatives on subprime MBS do not fall into this category, since they are zero-sum trades between investors and do not constitute real investment. (Although they do pay real financial transaction fees, part of the Consumption part of the equation).

  9. Looking at NGDP growth 2001-06 through the equation above, we see that the "Negative NPV Investment" constituted a significant part of that growth. This was unfortunate, but higher rates would have merely brought about our present situation seven years sooner. And even with the Negative NPV portion of Investment, the NGDP growth was not enough to sustain job growth.

    The ideal case is where we have sufficient NGDP growth 2001-06 AND have relatively little Negative NPV Investment, and the negative NPV investment we do have should have losses accrue to the investors.

    Without the Negative NPV Investment, the supply curve in the above interest rate model shifts to the left. Therefore, at the same interest rate, investment goes down and consumption stays the same. We would need lower real interest rates to increase both consumption and investment until the NGDP target is hit.

    The new investment is still Positive NPV investment because more investments make sense at low rates. Meanwhile, lower real rates mean that consumption now versus consumption later becomes more attractive (since all investment is eventually consumed). As a crazy, radical economist called Greg Mankiw argued in 2009, there is no honest reason nominal rates couldn't go negative if needed. All of the monetary base must either be in a Fed bank account or physical cash. If the Fed charges interest excess reserves and limits conversions of reserves to cash, banks would need to pass on negative rates on the interbank market and to customers. 0 is just another number.