Pages

Wednesday, June 14, 2017

Musings on June's FOMC Meeting

The FOMC decided today to raise its target interest rate so that it now sits in the 1.00-1.25 percent range. This move was largely expected and the FOMC continues to signal via its economic projections that it wants one more interest rate hike this year. Nothing terribly new here, but there were several developments today that caught my attention and are worth considering.

First, the FOMC released a surprisingly detailed plan of how it will unwind its balance sheet later this year. Fed chair Janet Yellen also said during the press conference these plans could be "put in effect relatively soon" if the data come in as expected. The announcement today can be seen as part of the FOMC's ongoing efforts to get the markets ready for the shrinking of its balance sheet. 

To shed light on this development, recall that the main theory the Fed used to justify the the large scale asset purchases was the portfolio channel. It says the Fed's purchase of safe assets would force investors to rebalance their portfolios toward riskier assets. This rebalancing, in turn, would reduce risk premiums, lower long-term interest rates, and push up asset prices. In turn, these developments would support the recovery.

A key step in this story was the Fed reducing the relative supply of safe assets to the public. One way to see it is through the growth of the Fed's share of marketable treasury securities. This can be seen in the figure below. 


In addition to the QE programs, the figure reveals a rather striking development. Since about September 2014 the Fed's share of marketable treasury securities has been falling. This has been a passive development for the Fed--it has maintained a fixed level of treasury holdings while total government debt has grown--but it is effectively QE in reverse. Per the portfolio channel, this reverse QE should have caused long-term treasury yields to rise. Instead, they have more or less been falling over this period:


Put differently, the Fed has been effectively shrinking its balance sheet for several years now and it has been much ado about nothing. Whatever influence QE had, its seems to be dwarfed by other economic forces driving long-term treasury yields.

It should not be surprising, then, that the Fed's announcement today about how it plans to shrink its balance sheet and Janet Yellen's follow-up comments did not create another taper tantrum. Yes, the Fed has been conditioning the market for the eventual reduction for several months, but maybe the bigger story here is that QE really did not have that big of an effect on interest rates and the recovery in the first place. Jim Hamilton reaches a similar conclusion:
[T]oday’s evidence seems to reinforce the conclusion that many have drawn about the effects of the Fed’s large-scale asset purchases–whatever effects these may have had on long-term interest rates were likely less important than other fundamentals. That appeared to be the case on the way to building up the Fed’s balance sheet, and so far appears to be the case in the long process of bringing the balance sheet back down.
To be clear, QE1 probably had a meaningful effect since it was applied in the midst of the financial crisis. However, I am becoming less convinced that QE2 and QE3 mattered much except for signaling purposes. As I have written elsewhere, the fundamental flaw with these programs was that they were beholden to the Fed's desire for rigidly low inflation and therefore consigned to be temporary monetary injections. Which leads me to the other development that really caught my attention.

Second, during the FOMC press conference, Fed chair Janet Yellen once again attributed the unexpectedly low inflation in recent months to one-off events.  She specifically attributed the below-target inflation to lower prices for cell phones and pharmaceutical. Here was my real-time response on twitter:


Yep, either the Fed is the most unlucky institution in the world or the Fed has a problem. I think the latter. The Fed appears to have begun having a problem with 2 percent inflation around the time of the Great Recession. This can be seen in the FOMC's summary of economic projections (SEPs) figure below. It shows for each FOMC meeting where SEPs were provided to the public the central tendency forecast of the core PCE inflation rate over the next year. The horizon for these forecasts depend on the time of the year they were released and range from one year to almost two years out. The forecast horizons are long enough, in other words, for the FOMC to have meaningful influence on the inflation rate.


The figure shows that since 2008, the FOMC has consistently forecasted at most 2 percent inflation. Note that The FOMC’s description of the SEP states (emphasis added) “Each participant's projections are based on his or her assessment of appropriate monetary policy. Longer-run projections represent each participant's assessment of the rate to which each variable would be expected to converge under appropriate monetary policy…” As former FOMC member Naranaya Kocherlakota notes, this means the projections reflect what members think inflation should be if they had complete control over monetary policy over the forecast horizon. It reveals their preferences for the future path of monetary policy. Consequently, the central tendency of FOMC members since 2008 indicates that they see the optimal inflation rate not at 2 percent, but at a range between 1 and 2  percent. 

The actual performance of the Fed's preferred inflation measure, the PCE deflator, has been consistent with this view. It has averaged about 1.5 percent since the recovery started in mid-2009. That is a revealed preference, not a series of accidents caused by bad  luck.

To be clear,  the Fed has only been explicitly targeting inflation at 2 percent since 2012, but many studies have shown it to be implicitly doing so since the 1990s. So this truly has been an eight-year plus problem for the Fed and one that makes Janet Yellen's remarks all the more disappointing to hear. One would think after almost a decade of undershooting 2 percent inflation there might be an acknowledgement from the FOMC like the one that came from Minneapolis Fed President Neel Kashkari (my bold):
[O]ver the past five years, 100 percent of the medium-term inflation forecasts (midpoints) in the FOMC’s Summary of Economic Projections have been too high: We keep predicting that inflation is around the corner. How can one explain the FOMC repeatedly making these one-sided errors? One-sided errors are indeed rational if the consequences are asymmetric. For example, if you are driving down the highway alongside a cliff, you will err by steering away from the cliff, because even one error in the other direction will cause you to fly over the cliff. In a monetary policy context, I believe the FOMC is doing the same thing: Based on our actions rather than our words, we are treating 2 percent as a ceiling rather than a target. I am not necessarily opposed to having an inflation ceiling...  I am opposed to stating we have a target but then behaving as though it were a ceiling.
It is time for the FOMC to come clean on what it really wants to do with inflation. Until it does so continue to expect confusion and  repeated questions to Fed officials over the Fed's inflation target.

5 comments:

  1. Great post.

    Kevin Erdmann over at Idiosyncratic Whisk notes that inflation minus shelter, food and energy is running at 0.6% YOY. Basically, property zoning has been propping up housing inflation, but that might have run its course. Also note China capital not entering US property as much due to Beijing capital controls. Vancouver Canada house prices down 25%.

    Spooky time.

    Seems the Fed could engineer a recession, and then be reluctant to admit it and change gears quickly.

    Also, I disagree a little on QE. I think open-ended QE 4 worked, and would have worked better if the Fed had gone bigger and, as you say, said it was permanent. Yeah, IOER is another puzzle.

    I wonder also about the "portfolio effect." Yes, but also sellers of Treasuries to the primary dealers can become consumers. For example, I sell $10k T bill to Nomura (who resells to Fed) and I buy a new car.

    With QE, no one in the private sector gives up present consumption to buy the $10k Treasury from me. Short-hand, the Fed prints the money and buys the Treasuries.

    The $10k Treasury seller (me) may re-invest proceeds, or buy a new car. Or wide-screen TV, baby, and lots of booze.


    ReplyDelete
    Replies
    1. Ben, to be clear, I am not saying QE had no effect. Studies that have looked at QE find on the high end that the Fed's QE programs lowered 10 year treasury yield about 100 bps. But the 10-year yield fell from a high of about 5.525% before the crisis to a low of about 1.4%. The Fed's part was not that important and that is assuming the high-end estimates are correct. If we move on to the real effect of the QE programs there is even less evidence that it mattered that much. Again, I do think QE1 probably mattered a lot and there was something to QE signaling the Fed's seriousness.

      Delete
    2. I'm not sure that changes in interest rates are the right way to measure the impact of the program. If it were really, really stimulative, then long term rates are more likely to rise, aren't they?

      Delete
  2. QE1 mattered to the banking system more than anything bc the asset swap from long duration bonds to cash + excess reserves liquified a frozen money market.

    ReplyDelete
  3. DB--Well, I guess I cannot prove that QE helped demand, though I think it did.

    But, if what you say is true, I come back to an interesting point. If QE is merely neutral in terms of demand and inflation…why not pay down the national debt?

    Why not have Fed do as they have done in Japan, and cut national debt in half?

    Even worse, what is the point in the Fed selling its balance sheet, and shoving $4 trillion of debt back onto taxpayer's backs?

    Side note: Love your podcasts, still looking forward to Adair Turner. That would be your all-time best podcast.

    ReplyDelete