Friday, January 25, 2013

Why We Need More Private Safe Assets: Risk Premiums, the Triffin Dilemma, and Cyclical Changes

Matthew C. Klein of the Economist and I have been debating how best to address the shortage of safe assets. His view is that fiscal policy alone can solve this problem by creating more safe assets. I, on the other hand, believe that fiscal policy cannot create enough assets to close the safe asset shortfall without jeopardizing the U.S.Treasury's risk-free status. The solution, then, lies with increasing investors' appetite for privately-created safe assets. Should this happen, financial firms would respond by creating the assets needed to close the safe asset gap.  

In his latest response, Klein argues that investors' appetite for private assets has already increased to no avail, that the government can create sufficient safe assets with jeopardizing its safe asset status, and that the main story here is most likely a secular shift in safe asset demand, not a cyclical one. I disagree on all counts and believe my approach--catalyzing private safe asset creation through an aggressive NGDP level target--is still the better one to take.  Let me explain why.

First, his claim about increased appetite for private assets ignores systematic evidence that shows the risk premium still needs to come down. It is true, as Klein points out, there has been record growth in junk bonds, but this is a small part of the debt market.  And yes, CLO issuance is picking up, but it is a pittance of what the market use to be as seen in this IMF figure. In other words, these are movements in the right directions, but the scale is small. If these were significant changes or harbingers of significant changes, risk premiums would be falling but they are not. Let me share two examples that demonstrate this fact. First, consider the risk premium indicated by the spread between Moody's BAA corporate yield and the 10-year treasury yield. Moody's BAA measure is a reflection of the entire corporate bond market at this lowest investment grade level. The figure below shows this spread remains elevated and according to the Survey of Forecasters, it is expected to remain elevated through early 2014.

And yes, this spread is systematically related to expected inflation and the stock market, so it matters. Consequently, a return of the risk premium to its average value would be associated with a pick-up in economic growth.

Another important measure is the equity risk premium. This shows the spread between the S&P 500 earnings yield and the 20-year treasury yield. It too is at an unusually high level as seen below:

This is an important metric for Ed Bradford, a bond trader, who recently noted the following about this risk premium measure:
The equity risk premium (EY - 30 year bond yield) at the end of 2011 rose above 6% and is currently approximately 4%... The recent 4-6% ERPs are some of the highest levels since 1971.  There was only a brief period in the late 1970s that had similarly high levels of risk premium.  It is notable that the late 1970s episode was also after a brutal bear market (1973-1974).  Most of the time the risk premium has been capped at 2%.  This initial yield differential between equities and fixed income imply much better returns going forward for equities and very attractive opportunities for traders to place pair trades between the two assets classes (equity vs US Treasury bonds)... Assuming corporate earnings continue to grow and we avoid a recession, I expect the risk premium to decrease significantly.
In short, the equity risk premium measure is currently too high relative to where it should be given long-run economic fundamentals. So again, there is room for the risk premium to drop and the public to increase its appetite for private assets. When this happens, it should catalyze private safe asset creation. 

Second, Klein believes that fiscal policy can almost costlessly create enough assets to close the safe asset shortage. As I noted in my last post, I think this is wrong because of the Triffin dilemma:
[T]he U.S. government faces a tension. It can run larger budget deficits to meet the global demand for safe assets, but doing so may eventually jeopardize its risk-free status, the very thing driving the demand for its securities. This is the modern version of the Triffin dilemma and it reminds us that there is a limit to how much safe asset creation can be done by the government.
This should not be a controversial position. Should investors start dumping treasuries because they fear an implicit default via higher expected inflation, the government would face higher financing costs. As expected inflation increased, the Fed would have to raise interest rates or allow higher inflation to emerge. We are nowhere near this point now, but how do we know that we will not hit it if fiscal policy tries to completely satiate the demand for safe assets?

Third, Klein's arguments seem to be premised on a view that what is driving the increased demand for safe assets is largely a change in secular demand, rather than a cyclical response to the economic downturn.  If so, then, the elevated risk premium would be the new norm. This is a reasonable point, as an aging global population, changes in accounting practices for pension funds, and the Asia-oil exporters' saving glut all put downward pressure on the term premium of  long-term treasury interest rates. The problem, though, is that these developments were all happening well before 2007, the period when long-term treasury yields began their now 5-year decline. Something  very different began in 2007 that pushed long-term treasury yields on a new downward trajectory. This sharp break is very clear in the figure below:

It is also clear in nominal treasury yields. And, as everyone knows, what began in 2007 was the economic crisis from which we have not fully recovered. The crisis has lowered both the expected path of short-term interest rates (because of ongoing expected economic weakness) and the term premium as investor flock to the safety of treasuries. A robust recovery, where investors returned to private safe assets, should reverse both of these developments and push yields closer to their pre-crisis values. This cyclically-driven drop is yields is not that different than the one during the Great Depression. Then too, yields were depressed for a long time since the economy was depressed for a long time.

For these reasons, we need a NGDP level target now more than ever.

P.S. Thanks to Ed Bradford for the Bloomberg figures.  

Update: I originally had the below picture in the post, but decided it would be better down here. It is former gubernatorial candidate Jimmy McMillan agreeing with me:


  1. Why do we *need* more safe assets? Why don't we just let the existing ones rise in value, thereby providing safety?

    If we wanted to express our desire for safety by buying fire extinguishers, then I'd agree that we need to produce more safe assets. After all, only some sort of increase in the supply of extinguishers will be able to meet that demand.

    But things are different if we express our demand for safety by turning to financial markets. The great thing about t-bonds is that unlike fire extinguishers, we don't need to fabricate more of them to meet our demands for safety... we just need a higher real value on the stock of existing t-bonds. This can be entirely met by shifts in prices. Where is the problem that needs to be rectified?

    (I know I've asked you these things before. Maybe you'll get through to me this time)

    1. Great question. I agree that if treasury prices adjusted enough it would take of the safe asset shortage. But since we are living near the ZLB that can't happen. Interest rates would need to go below zero, but they won't since investors could earn at least 0% holding money. Thus, treasuries can't increase enough in value to clear the market.

      Also, since this clearing is not happening we still have an excess money demand problem. Which means the price level will not adjust very quickly to clear the market.

      Does that make sense?

    2. Yes, it makes sense.

      But I'm looking at US treasury yields. The 4.125% bond maturing on 5/15/2015 is priced at 108.62. But taking out my handy dandy bond calculator, it can rise in price to 109.63 before it trades at a 0% yield. If there was an unmet demand for safe assets, why is the 5/15/2015 4.125% bond only trading at 108.62 and not 109.63?

  2. Isn't there a good argument to be made that there are too many safe assets?

    1. Dan, that sounds intriguing could you explain it a bit more?

  3. From Gofx. Per my comment on the last thread, hedging etc. can "make" assets safer. Thus a private creation of more "safe" assets can occur. There is a market price for such insurance. No asset can be completely safe. You can have counterparty risk on a hedge/derivative, but you can also have counterparty risk in terms of a Sovereign and associated default or political risk.

  4. Dan Kervick hits the ball out of the stadium. The problem is indeed too much safety - we need to be encouraging entrepreneurship and risk-taking, not providing featherbedding for the 1%. Our GDP has been stagnating for a decade due to this misplaced strategy.

    1. Anonymous, if the risk premium was lowered that is exactly what would happen.

  5. I don't think you can compare pre and during QE "risk" premiums involving treasuries. QE is selectively lowering the yield on treasuries, not on BAA corporate bonds or equities. According to Bernanke ( ), the effect is about 80-120bps at ten years - about the same as your excess risk premium on Baa corporates. The "break" downward in your chart of treasury yields looks more like 2009 than 2007 to me.