Archive for the Category Market monetarism

 
 

Another Market Monetarist Advisory

A few weeks ago I alerted readers to NGDP Advisers, which features Marcus Nunes, James Alexander, Benjamin Cole and Justin Irving.  Now we are about to see another MM advisory firm.  Lars Christensen will launch Markets and Money Advisory early next year.  Lars has a new post on the Riksbank, which provides an example of the sort of analysis he will be doing:

Believe it or not – there is a country in the world where I now believe that monetary policy is becoming (moderately) too easy. Yes, that is correct – I will not always say that monetary policy is too tight. The country I talk about is Sweden. More on that below.

Assessing monetary conditions

I strongly believe that the assessment of the monetary stance of a country should not be based on for example looking at the level of nominal interest rates, but rather on whether or not the country is on track to hitting the central bank’s nominal target in lets say 12-18 months.

A way of assessing that is of course to look at market inflation expectations (if the central bank targets inflation as in the case of Sweden’s Riksbank). If inflation expectations are below (above) the target (for example 2%) then monetary conditions are too tight (easy).

An alternative to this approach is to look at other monetary indicators – for example money supply growth, nominal GDP growth, interest rates and the exchange rate. And this is exactly what we are doing in our (Markets & Money Advisory’s) upcoming publication on Global Monetary Conditions.

Policy consistency  

Hence for all of the nearly 30 country we analyse in the publication we look at the four monetary indicators mentioned above and compare the development in these indicators with what we believe would be consistent with the given central bank’s inflation target.

I don’t know enough about Sweden to comment, but it certainly is an interesting case.  Here’s Bloomberg:

As the Brits worry about the ramifications of a weakening pound, Sweden’s central bank has happily driven down its currency to the lowest level in more than half a decade.

Defying fundamentals – strong economic growth, a big current account and trade surplus and rising employment – the Swedish krona was the second-worst performing currency in the world last month after the British pound.

Those “fundamentals” may be interesting, but they are not the sort of fundamentals that a central bank should focus on.  Instead, inflation and NGDP growth are what matter.  Just because you have a big current account surplus does not mean that money is too easy.  Indeed Japan experienced both deflation and a CA surplus at the same time.  Sweden’s surplus merely reflects its high saving rate; it tells us nothing about whether the exchange rate is at the wrong level.  For that, you need to look at NGDP growth.  Thus the Japanese yen is too strong because NGDP growth is too slow.  Lars suggests that Sweden’s NGDP growth is excessive, and that’s why he thinks they are too easy right now.

Sweden has other important lessons.  Compared to Denmark (4.2%) and Norway (5.0%), Sweden has a rather high unemployment rate–currently 6.6%.  But that probably reflects a higher natural rate of unemployment, which cannot be fixed with monetary policy.  Again, monetary policy cannot be used to “solve problems”—instead it should aim at steady NGDP growth rates to avoid creating problems.  But don’t go to the other extreme and “oppose monetary policy”.  There is no such things as not using monetary policy, and any attempt to refrain from using it will merely create bad (highly unstable) monetary policy.  I say this because I’m seeing this mistake more and more often.  This headline made me cringe:

Buiter: Forget Monetary Policy, It’s Had Its Day

Yes, and eating food won’t solve your problems, so STOP EATING FOOD YOU IDIOT!

Or perhaps I should say. “You may not care about monetary policy, but monetary policy cares about you.”

PS.  The Bloomberg article on Sweden suggests that current policy is turning the krona into “play money”.  That might be a bit strong, given that Sweden’s inflation rate is currently 1%.  That’s a tad below Zimbabwe 2008 levels.

Bernanke on the Fed’s new view of the economy

Ben Bernanke has a post discussing the Fed’s evolving view of the economy:

I’ll focus here on FOMC participants’ longer-run projections of three variables—output growth, the unemployment rate, and the policy interest rate (the federal funds rate)—and designate these longer-run values by y*, u*, and r*, respectively. Under the interpretation that these projections equal participants’ estimates of steady-state values, each of these variables is of fundamental importance for thinking about the behavior of the economy:

  • Projections of y* can be thought of as estimates of potential output growth, that is, the economy’s attainable rate of growth in the long run when resources are fully utilized

  • Projections of u* can be viewed as estimates of the “natural” rate of unemployment, the rate of unemployment that can be sustained in the long run without generating inflationary or deflationary pressures

  • Projections of r* can be interpreted as estimates of the “terminal” or “neutral” federal funds rate, the level of the funds rate consistent with stable, noninflationary growth in the longer term

He then explain how over the past few years the Fed has tended to consistently overestimate these variables:

Why are views shifting?  The changing views of FOMC participants (and of most outside economists) follow pretty directly from persistent errors in forecasting economic developments in recent years:As the table shows, FOMC participants have been shifting down their estimates of all three variables—y*, u*, and r*—for some years now.

Notice that there is no mention of the fact that the markets, and hence market monetarists, have generally been more accurate than the Fed.  We take financial market predictions seriously, and thus immediately discounted the Fed forecast of 4 rate increases in 2016, made back in December.  Indeed for years I’ve been arguing that the Fed’s dot plot is too optimistic about the Fed’s ability to raise interest rates under its current policy regime.

More than two years ago I suggested that 3% NGDP growth and 1.2% RGDP growth were the new normal, at a time when the Fed was still forecasting considerably higher rates.  Bernanke says the lower natural GDP growth rate is partly due to surprisingly low productivity growth.  Back in 2011, I suggested that we were having a “job-filled non-recovery“, just the opposite of the jobless recovery being discussed by many pundits.  Since then we’ve continued to have a job-filled non-recovery, with faster that expected job creation and a fast falling unemployment rate, accompanied by slower than expected RGDP growth.  This is just another way of saying that productivity growth has been lousy (indeed negative for three quarters in a row.)

It’s nice that the Fed is finally seeing the light on issues that market monetarists have been emphasizing for many years.  But I’d feel better if they took this as a lesson that they need to change their entire operating system, and start relying much more on market forecasts.

Back in 1997, Bernanke published a paper with Michael Woodford (in the JMCB) suggesting that market forecasts could be useful to policymakers, if they revealed information about the impact of different monetary policy instrument settings.  OK, so why doesn’t the Fed take Bernanke’s advice and create a set of prediction markets for inflation and output, one for each plausible instrument setting.

PS.  By “job-filled non-recovery” I did not mean that we were not getting closer to the natural rate, I meant we are not recovering in the sense of going back to the old trend line.  Clearly the labor market has been gradually recovering.

HT:  Bill Beach, Patrick Horan

 

Tim Duy discusses the evolving views of Fed Governor Powell

Tim Duy has a very good new post, showing how Jerome Powell is moving in a more dovish direction.  The following quotation is Powell, with the remark about the flat Phillips curve being Tim:

When I was first exposed to macroeconomics in college, more than four decades ago, the view was that inflation was strongly influenced by the amount of slack in the economy. But the relationship between slack and inflation has weakened substantially over the years.

Or, in other words, the Phillips Curve is flat. Not quite flat as a pancake, but pretty darn flat. More important:

In addition, inflation depends importantly on the inflation expectations of workers and firms. A widely shared view among economists today is that, unlike during the 1970s, expectations are no longer heavily influenced by fluctuations in inflation, but are fairly constant, or anchored. For both these reasons, inflation has become less responsive to cyclical changes in the economy.

I’d go even further.  The Phillips curve is not useful because it is NGDP, not inflation, that best explains how nominal and real variables are related. And the causation goes from the nominal to the real (NGDP to unemployment) not the real to the nominal (unemployment to inflation).

Once again, here’s Powell, with a follow-up comment by Duy:

I am often asked why rates remain so low now that we are near full employment. A big part of the answer is that, at least for the time being, the appropriate level of rates is simply lower than it was before the crisis. As a result, policy is not as stimulative as it might appear to be. Estimates of the real interest rate needed to keep the economy on an even keel if it were operating at 2 percent inflation and full employment–the “neutral rate” of interest–are currently around zero. Today, the real short term interest rate is about negative 1-1/4 percent, so policy is actually only moderately stimulative. I anticipate that the neutral rate will move up over time, as some of the headwinds that have weighed on economic growth ease.

The Fed increasingly recognizes that policy is not highly accommodative simply because rates are zero. The stance of policy is relative to the real interest rate, and a lower real rate means that policy is actually only “moderately” stimulative. Translation: There is no need to hike rates soon because policy is not particularly accommodative.

Of course market monetarists have been saying  that low rates don’t mean accommodative policy ever since 2008.  I’d go even further.  Not only is the current policy not as accommodative as it seems, it’s not accommodative at all.  NGDP growth (or inflation) are likely to undershoot the Fed’s goals.

Over the period since 2009, we’ve seen macroeconomic discourse evolve as follows:

1.  NGDP may be a more useful indicator of nominal conditions than inflation.

2.  The Phillips Curve is not very useful.

3.  Low interest rates do not imply that money is easy.

4.  Expansionary fiscal policy may be offset by an inflation targeting central bank.

5.  The zero lower bound does not prevent negative IOR.

6.  At the zero bound, a premature increase in interest rates will lead to lower interest rates in the long run.

Of course no one has a monopoly on these views, but which set of bloggers were most forcefully making these points in early 2009?

With the post-Brexit vote plunge in global bond yields, any doubts that low rates are the new normal are gone.  The Fed’s been much slower than the markets to understand this new reality, but they aren’t stupid.  At some point the Fed will realize that its preferred (“conventional”) policy tool simply doesn’t work.  Rates will immediately fall to zero in all future recessions, so “conventional” monetary policy will be useless.  How will the Fed react:

1.  NGDP targeting

2.  A higher inflation target

3.  Level targeting

4.  Miles Kimball’s negative IOR plan

5.  Throw up their hands and ask for support from fiscal policy

I hope and pray they don’t choose option #5. Because it won’t work.

PS.  Tyler Cowen just reported that Swiss yields are negative out to 50 years.  That’s why I opposed the Swiss decision to revalue the franc upward last year.  The upward revaluation was motived by a fear of inflation (and no, I’m not kidding.)

HT:  David Levey

Greg Ip on monetary policy

Where does all the time go?

I just noticed that I’ve fallen behind on the set of podcasts by David Beckworth, so I will work through the ones I’ve missed, starting with the Greg Ip, one of our best economic journalists.  Here’s my favorite comment by Ip:

And it actually may be better to have lots of small financial disruptions than one big financial disruption.

In Greg’s recent book he discusses this idea in more detail.  In the interview, Greg uses analogies such as the danger of continually preventing small forest fires, and thus building up fuel for a catastrophic fire.

Over the past 50 years the government has prevented financial crises about every decade or so, by either bailing out depositors of large banks, or arranging assistance in the case of LTFC (1998).  And this had the effect of storing up fuel (moral hazard) for an even bigger crisis in 2008.  But I would go much further that Ip, who approves of FDIC.  It’s not politically possible to abolish FDIC, but perhaps we could create a two-tier system where insured deposits are backed by safe assets, so that taxpayers are not put at risk.  Deposits used for lending to businesses and homebuyers would not be insured, but would offer higher interest rates to depositors.  Let bank depositors choose how much risk they are willing to take.  Ip also is appropriately critical of the regulatory overreach of Dodd-Frank. BTW, banks would hate my FDIC reform proposal, but it could be combined with the complete repeal of Dodd-Frank.

There are also a few areas where I disagreed with Ip.  At one point he wondered why there was so much discussion of the need for monetary stimulus. After all, unemployment in the US and Japan is relatively low, and the unemployment rate in the eurozone is now declining at a decent clip.  This is a good argument, but I think he’s also missing something important.  Monetary policy must be judged as a regime, not in terms of day-to-day considerations of macroeconomic stability.

For better or worse, central banks now focus most of their effort on inflation targeting, with some attention also paid to keeping unemployment close to the natural rate. Recall that the natural rate hypothesis predicts that the public will eventually adjust their expectations to match any inflation rate, and unemployment will eventually move back to the natural rate.  When viewed from this perspective, I think what Greg’s really asking is what difference does it make if the Eurozone has 1% inflation, or 1.9% inflation, as long as it is reasonably steady and as long as unemployment seems to be adjusting back to the natural rate.

I see two problems with the ECB allowing 1% inflation to be the new normal:

1.  If this were to occur, the public would lose faith in the ECB’s inflation promises. This would make ECB policy less effective in the next crisis.  If central banks are going to set inflation targets, then those targets should mean something.  If they decide not to target inflation (as I’d prefer) then it’s essential that they set some other target, such as NGDPLT.

2.  If 1.0% inflation, rather than 1.9% inflation, becomes the new normal in the ECB, then nominal interest rates will move to a permanently lower track.  And since real interest rates seem to be entering a new normal which is well below the rates we saw in the 20th century, a lower trend rate of inflation would mean that the ECB will be stuck at the zero bound for a much greater percentage of the time.  Indeed financial markets are already quite pessimistic about the future course of eurozone rates, especially for safe assets like German and Swiss long-term bonds.

Notice that points 1 and 2 relate to each other; both make it more difficult for the ECB to achieve its goals in the future.  So there is real value in taking an announced inflation target seriously, and trying to hit it.  BTW, the US is doing much better than the eurozone and Japan on the inflation front, but just today Kocherlakota warned that even the US is likely to fall short of 2% inflation going forward.  (I’m a moderate on this question—I think they’ll probably fall a bit short, but perhaps not as much as Kocherlakota and some of my fellow MMs believe.  I see something around 1.8% as the new normal.)

At one point Ip asked David the question of what should the Fed actually do to implement an NGDPLT policy regime.  I hate these “concrete steppes” questions, but we need to face the fact that this is what everyone wants to know.  The reason why I hate these questions is because I know the sort of answer people are looking for:

Desired answer:  Some big bazooka of a monetary policy instrument that is so powerful that it can clearly move NGDP to the desired policy path.

My answer:  NGDPLT is the big bazooka, and once implemented you merely need to do tiny little OMOs, like we did back before 2008.

And I know that this answer won’t satisfy anyone.  They think money has been very easy, and if we’ve fallen short then we must need very, very, very easy policy.  We MMs think money has been tight, and that a NGDPLT target could be hit with the sort of moderate policy we had before 2008.  In other words, if 5% NGDPLT were adopted in 2007, or right now, the policy would look pretty much like what you saw in Australia after 2007, or what you see in Australia today.  Positive interest rates.

But yes, you do need a big “instrument” bazooka lurking in the background, just in case.  That makes the system credible, so that you don’t actually have to use it. For David the big bazooka is the Treasury, promising to do a coordinated fiscal/monetary expansion if the Fed runs out of ammo.  For me the big bazooka is a Fed promise to buy any and all financial assets, anywhere in the world, until market expectations of NGDP growth are equal to 5% (or whatever the target chosen.)

And the other point I always make is that the lower the NGDP target (i.e. the lower the trend inflation rate) the bigger the Fed balance sheet as a share of GDP.  If NGDP growth is so low that nominal rates fall to zero, then the Fed balance sheet can get very large.  If the NGDP target rate is set high enough where rates stay above zero, then the Fed balance sheet stays small.  I prefer a small Fed balance sheet.

Inflation or socialism?  It’s your choice.

David Beckworth on EconTalk

David Beckworth was interviewed by Russ Roberts this morning in a special video version of EconTalk, which was held at the Cato Institute and hosted by George Selgin. Unfortunately I am not able to find a link for the talk, but I’ll put one up if someone else can direct me to it.

Update:  Here’s the link:

http://www.cato.org/events/econtalk-live-david-beckworth-monetary-policy-great-recession

Update#2:  I’m told the link no longer works, but a new link should be up in about 10 days.

There was some discussion of whether the market monetarist critique of Fed policy in 2008 is just Monday morning quarterbacking.  David seemed to concede that this complaint had some merit, and perhaps to some extent it does.  But I also think David is being too modest.  Here’s David criticizing interest on reserves, way back in October 2008, right after it was first adopted:

Is the Federal Reserve (Fed) making a similar mistake to the one it made in 1936-1937? If you recall, the Fed during this time doubled the required reserve ratio under the mistaken belief that it would reign in what appeared to be an inordinate buildup of excess reserves. The Fed was concerned these funds could lead to excessive credit growth in the future and decided to act preemptively. What the Fed failed to consider was that the unusually large buildup of excess reserves was the result of banks insuring themselves against a replay of the 1930-1933 banking panics. So when the Fed increased the reserve requirements, the banks responded by cutting down on loans to maintain their precautionary level of excess reserves. As a result, the money multiplier dropped and the money supply growth stalled as seen in the figure below.

.  .  .

Now in 2008 the Fed did not suddenly increased reserve requirements, but it did just start paying interest on excess reserves. The Fed, then, just as it did in 1936-1937 has increased the incentive for banks to hold more excess reserves. As a result, there has been a similar decline in the money multiplier and the broader money supply (as measured by MZM) which I documented yesterday. If the Fed’s goal is to stabilize the economy, then this policy move appears as counterproductive as was the reserve requirement increase in 1936-1937.

David says that in retrospect he thinks that Fed policy went off course even earlier, in the middle of 2008. Speaking for myself, I didn’t really become aware of the problems with monetary policy until September 2008, after the Fed refused to cut rates despite plunging TIPS spreads.  In retrospect, money became much too tight a couple of months earlier.

Because of data lags, it’s not always possible to predict a recession until the recession is already well underway.  During the past three recessions, a consensus of economists didn’t predict a recession until about 6 months in.  That’s right, not only are macroeconomists unable to forecast, we are also unable to nowcast.

This is why NGDPLT is so important. Under a level targeting regime, the market tends to prevent sharp drops in NGDP from occurring in the first place.  Under NGDPLT, the economy would not have fallen as sharply in late 2008, partly because level targeting would have prevented a steep plunge in asset prices, and partly because current AD is heavily dependent on future expected AD.  If you keep future expected AD rising along a 5% growth path, current AD will not fall very far during a banking crisis.

There was also some discussion of the shortage of safe assets.  Two questions came to mind:

1.  Does this theory imply that risk spreads should have widened in recent years, as the demand for T-bonds has increased faster than the demand for riskier bonds?

2.  Has this in fact occurred, and if so to what extent?