Must-Read: Pedro Bordalo, Nicola Gennaioli, and Andrei Shleifer: Diagnostic Expectations and Credit Cycles

Must-Read: Very clever indeed…

Pedro Bordalo, Nicola Gennaioli, and Andrei Shleifer: Diagnostic Expectations and Credit Cycles: “We present a model of credit cycles arising from diagnostic expectations…

…a belief formation mechanism based on Kahneman and Tversky’s (1972) representativeness heuristic. In this formulation, when forming their beliefs agents overweight future outcomes that have become more likely in light of incoming data. The model reconciles extrapolation and neglect of risk in a unified framework. Diagnostic expectations are forward looking, and as such are immune to the Lucas critique and nest rational expectations as a special case. In our model of credit cycles, credit spreads are excessively volatile, over-react to news, and are subject to predictable reversals. These dynamics can account for several features of credit cycles and macroeconomic volatility.

Must-read: Narayana Kocherlakota: “Information in Inflation Breakevens about Fed Credibility”

Must-Read: Whenever I look at a graph like this, I think: “Doesn’t this graph tell me that the last two years were the wrong time to give up sniffing glue the zero interest-rate policy”? Anyone? Anyone? Bueller?

Graph 3 Month Treasury Bill Secondary Market Rate FRED St Louis Fed

And Narayana Kocherlakota agrees, and makes the case:

Narayana Kocherlakota: Information in Inflation Breakevens about Fed Credibility: “The Federal Open Market Committee has been gradually tightening monetary policy since mid-2013…

…Concurrent with the Fed’s actions, five year-five year forward  inflation breakevens have declined by almost a full percentage point since mid-2014.  I’ve been concerned about this decline for some time (as an FOMC member, I dissented from Committee actions in October and December 2014 exactly because of this concern).  In this post, I explain why I see a decline in inflation breakevens as being a very worrisome signal about the FOMC’s credibility (which I define to be investor/public confidence in the Fed’s ability and/or willingness to achieve its mandated objectives over an extended period of time).

First, terminology.   The ten-year breakeven refers to the difference in yields between a standard (nominal) 10-year Treasury and an inflation-protected 10-year Treasury (called TIPS).  Intuitively, this difference in yields is shaped by investors’ beliefs about inflation over the next ten years.  The five-year breakeven is the same thing, except that it’s over five years, rather than 10.  

Then, the five-year five-year forward breakeven is defined to be the difference between the 10-year breakeven and the five-year breakeven.   Intuitively, this difference in yields is shaped by beliefs about inflation over a five year horizon that starts five years from now.   In particular, there is no reason for beliefs about inflation over, say, the next couple years to affect the five-year five-year forward breakeven. 

Conceptually, the five-year five-year forward breakeven can be thought of as the sum of two components:
 
1. investors’ best forecast about what inflation will average 5 to 10 years from now

  1. the inflation risk premium over a horizon five to ten years from now – that is, the extra yield over that horizon that investors demand for bearing the inflation risk embedded in standard Treasuries.
     
    (There’s also a liquidity-premium component, but movements in this component have not been all that important in the past two years.) 

There is often a lot of discussion about how to divide a given change in breakevens in these two components.  My own assessment is that both components have declined.  But my main point will be a decline in either component is a troubling signal about FOMC credibility.  

It is well-understood why a decline in the first component should be seen as problematic for FOMC credibility.  The FOMC has pledged to deliver 2% inflation over the long run.  If investors see this pledge as credible, their best forecast of inflation over five to ten year horizon should also be 2%.   A decline in the first component of breakevens signals a decline in this form of credibility.  

Let me turn then to the inflation-risk premium (which is generally thought to move around a lot more than inflation forecasts). A decline in the inflation risk premium means that investors are demanding less compensation (in terms of yield) for bearing inflation risk. In other words, they increasingly see standard Treasuries as being a better hedge against macroeconomic risks than TIPs.  

But Treasuries are only a better hedge than TIPs against macroeconomic risk if inflation turns out to be low when economic activity turns out to be low.  This observation is why a decline in the inflation risk premium has information about FOMC credibility.  The decline reflects investors’ assigning increasing probability to a scenario in which inflation is low over an extended period at the same time that employment is low – that is, increasing probability to a scenario in which both employment and prices are too low relative to the FOMC’s goals. 

Should we see such a change in investor beliefs since mid-2014 as being ‘crazy’ or ‘irrational’? The FOMC is continuing to tighten monetary policy in the face of marked disinflationary pressures, including those from commodity price declines.  Through these actions, the Committee is communicating an aversion to the use of its primary monetary policy tools: extraordinarily low interest rates and large assetholdings. Isn’t it natural, given this communication, that investors would increasingly put weight on the possibility of an extended period in which prices and employment are too low relative to the FOMC’s goals?

To sum up: we’ve seen a marked decline in the five year-five year forward inflation breakevens since mid-2014.  This decline is likely attributable to a simultaneous fall in investors’ forecasts of future inflation and to a fall in the inflation risk premium.   My main point is that both of these changes suggest that there has been a decline in the FOMC’s credibility.   

To be clear: as I well know, in the world of policymaking, no signal comes without noise.  But the risks for monetary policymakers associated with a slippage in the inflation anchor are considerable.   Given these risks, I do believe that it would be wise for the Committee to be responsive to the ongoing decline in inflation breakevens by reversing course on its current tightening path.

Rethinking the expectations channel of monetary policy

From the “Meeting Report” section of the Fall 2015 Brookings Papers on Economic Activity:

”Fredric Mishkin elaborated on the issues that discussant Adam Posen had raised…

…regarding how demoralizing the outcomes from Japanese monetary policy have been. He had felt more strongly than Posen that expectations were very important and that managing expectations is a key element in good monetary policy. He and his colleagues expected much stronger effects in Japan from the expansion of its monetary policy. Japan’s outcome might demonstrate that raising inflation expectations is much more difficult than lowering them, and moreover this might be true globally.

Acknowledging that he is known to be a big proponent of inflation targeting, Mishkin said that when the focus is on how to keep inflation expectations down, it has worked well. But he and others have found it much more difficult to raise expectations, particularly during a long period of deflation. He noted that the general view about New Zealand’s experience is that the policy there anchored inflation expectations, that it had an expansionary impact by raising inflation expectations and thereby getting people to spend more. Yet one can see how difficult it has been to get the Japanese consumer to do the same…

”Brad DeLong seconded Mishkin’s comment…

adding that the macroeconomic situation in Japan has not developed necessarily to Japan’s advantage even though economists had strong reasons to think the expectations channel was present based on historical examples. In the 1930s, Franklin Roosevelt’s New Deal and Neville Chamberlain’s announcement as Chancellor of the Exchequer that his policy was to restore Great Britain’s price level to its pre-Depression state both demonstrated the power of the expectations channel. Indeed, the same happened when Japanese Finance Minister Takahashi Korekiyo announced his decision to go for reflation in Japan in the 1930s. It is a great puzzle that this time around it has not been working…

Indeed.

Add the failure of Abenomics to perform as expected–at least, as I had expected–to my list of major career analytical howlers, along with:

  • Central banks have the power and the will to return North Atlantic nominal GDP to its pre-2008 growth path…
  • Subprime is too small a market to be a major source of systemic risk…
  • Repealing Glass-Steagal would lead to a more efficient and less lavish and disruptive Wall Street…
  • NAFTA will tend to strengthen the peso, as capital is now attracted by guaranteed tariff- and quota-free access to the largest consumer market in the world…

The failure of Abenomics to perform as expected may well be the analytical error that is the sign of the deepest and most significant flaw in my Visualization of the Cosmic All. Ever since I was 20 or so I have operated with the rules of thumb that labor-market expectations are by and large backward-looking well financial-market expectations are by and large forward-looking. These rules of thumb have served me relatively well in the past. But they are not serving me well today.

Yes, I have long known that there is a noise trader term in asset prices. But I’ve always taken it to be a persistent additive deviation from the forward-looking expectational equilibrium price fundamental. It now seems that this is not good enough to help me properly understand the world. But with what should I replace it?

Weekend reading: Narayana Kocherlakota: “Dovish Actions Require Dovish Talk (To Be Effective)”

The wise Mark Thoma sends us to the newly-unmuzzled and very sharp Narayana Kocherlakota: Dovish Actions Require Dovish Talk (To Be Effective): “The Federal Open Market Committee (FOMC)…

…has bought a lot of assets and kept interest rates extraordinarily low for the past eight years.  Yet, all of this stimulus has accomplished surprisingly little (for example, inflation and inflation expectations remain below target and are expected to do so for years to come).   Does that experience mean that we should give up on monetary policy as a useful way to stimulate aggregate demand?
My answer is no.  I argue that, over the past seven years, the FOMC’s has consistently talked hawkish while acting dovish.  This communications approach has weakened the effectiveness of policy choices, probably in a significant way.  Future monetary policy stimulus can be considerably more effective if the FOMC is much more transparent about its willingness to support the economy – that is, about its true dovishness.

My starting point is that households and businesses don’t make their decisions about spending based on the current fed funds rate – which, is after all, a one-day interest rate.  Rather, spending decisions are based on longer-term yields.  Those longer-term yields depend on market participants’ beliefs about how monetary policy will evolve over the next few years.  Those beliefs are a product of both FOMC actions and FOMC communications. 

In December 2008, the FOMC lowered the fed funds rate target range to 0 to a quarter percent. It did not raise the target range until December 2015, when the unemployment rate had fallen back down to 5%.   But – with the benefit of hindsight – a shocking amount of this eight years’ worth of unprecedented stimulus was wasted, because it was largely unanticipated by financial markets. (Full disclosure: I took part in FOMC meetings from November 2009 through October 2015, and it could certainly be argued that I was part of the problem that I describe until September 2012.)  

I’ll illustrate my basic point in the most extreme way that I can.  In November 2009, the Committee’s statement said that the fed funds rate might be raised after ‘an extended period’ – a term that was generally interpreted to mean ‘about six months’.  Accordingly, as footnote 25 of this speech notes, private forecasters in the Blue Chip survey projected that the unemployment rate would be near 10 percent at the time of the first interest rate increase.  

Now, suppose that the FOMC had communicated its true reaction function in November 2009 (or even as late as December 2012): as long as inflation was anticipated to be below 2% over the medium-term, the Committee would not raise the fed funds rate until the unemployment rate had fallen to 5% or below.  We can’t know the impact of such communication with certainty.   But most macroeconomic models would predict that this kind of statement would have put significant upward pressure on employment and prices.  In other words: the models predict that if the FOMC had been willing to communicate its true willingness to support the economy, the Committee would have been able to (safely) raise rates much sooner.  

I want to be clear: my point in this post is not to express regrets or recrimination over past ‘mistakes’.    (It would have been good in 2009 to know what we know now, but we didn’t.)  And my point is not that monetary policy is some kind of panacea.  In the presence of a lower bound on nominal interest rates, expansionist fiscal policy would have been helpful in the past (and could be now too). 

My point is this: we shouldn’t make judgements about the efficacy of future monetary policy stimulus based on the experience from the past seven years.   Unfortunately, much of the potential impact of that lengthy stimulus campaign was vitiated by the FOMC’s generally hawkish communications.   

In my view, the FOMC can deliver useful impetus to aggregate demand with its remaining tools.  But it needs to communicate ahead of time about its true willingness and ability to support the economy.   Without that prior communication, later attempts at stimulus are likely to prove in vain – and the Fed’s credibility may suffer further damage.

Must-read: Larry Summers: “Heed the Fears of the Financial Markets”

Must-Read: Larry Summers: Heed the Fears of the Financial Markets: “They understood the gravity of the 2008 crisis well before the Federal Reserve…

…Markets are more volatile than the fundamentals they seek to assess. Economist Paul Samuelson quipped 50 years ago, ‘the stock market has predicted nine of the last five recessions’. Former Treasury secretary Robert Rubin was right when he would regularly reassure anxious politicos in the Clinton White House that ‘markets go up, markets go down’ on days when a market move created either joy or anxiety…. Still, since markets are constantly assessing the future and aggregate the views of a huge number of participants, they often give valuable warning when conditions change…. Policymakers who dismiss market moves as reflecting mere speculation often make a serious mistake. Markets understood the gravity of the 2008 crisis well before the Federal Reserve. They grasped the unsustainability of fixed exchange rates in Britain, the UK, Mexico and Brazil while the authorities were still in denial, and saw slowdown or recession well before forecasters in countless downturns….

Signals should be taken seriously when they are long lasting and coming from many markets, as with current market indications that inflation will not reach target levels within a decade in the US, Europe or Japan…. It is conceivable that Chinese developments reflect market psychology and clumsy policy responses, and that the response of world markets is an example of transient contagion. But I doubt it. Over the past year, about 20 per cent of China’s growth as reported in its official statistics has come from its financial services sector…. This is hardly a case of healthy or sustainable growth…. Traditionally, international developments have had only a limited effect on the US and European economies because they could be offset by monetary policy actions…. Because of China’s scale, its potential volatility and the limited room for conventional monetary manoeuvres, the global risk to domestic economic performance in the US, Europe and many emerging markets is as great as at any time I can remember. Policymakers should hope for the best and plan for the worst.