I am increasingly realising that a key problem in the Market Monetarist arguments for NGDP level targeting is that we have not been very clear in our arguments concerning how it would actually work.
We argue that we should target a certain level for NGDP and then it seems like we just expect it too happen more or less by itself. Yes, we argue that the central bank should control the money base to achieve this target and this could done with the use of NGDP futures. However, I still think that we need to be even clearer on this point.
Therefore, we really need a Market Monetarist theory of the monetary transmission mechanism. In this post I will try to sketch such a theory.
Combining “old monetarist” insights with rational expectations
The historical debate between “old” keynesians and “old” monetarists played out in the late 1960s and the 1970s basically was centre around the IS/LM model.
The debate about the IS/LM model was both empirical and theoretical. On the hand keynesians and monetarists where debating the how large the interest rate elasticity was of money and investments respectively. Hence, it was more or less a debate about the slope of the IS and LM curves. In much of especially Milton Friedman writings he seems to accept the overall IS/LM framework. This is something that really frustrates me with much of Friedman’s work on the transmission mechanism and other monetarists also criticized Friedman for this. Particularly Karl Brunner and Allan Meltzer were critical of “Friedman’s monetary framework” and for his “compromises” with the keynesians on the IS/LM model.
Brunner and Meltzer instead suggested an alternative to the IS/LM model. In my view Brunner and Meltzer provides numerous important insights to the monetary transmission mechanism, but it often becomes unduly complicated in my view as their points really are relatively simple and straight forward.
At the core of the Brunner-Meltzer critique of the IS/LM model is that there only are two assets in the IS/LM model – basically money and bonds and if more assets are included in the model such as equities and real estate then the conclusions drawn from the model will be drastically different from the standard IS/LM model. It is especially notable that the “liquidity trap” argument breaks down totally when more than two assets are included in the model. This obviously also is key to the Market Monetarist arguments against the existence of the liquidity trap.
This mean that monetary policy not only works via the bond market (in fact the money market). In fact we could easily imagine a theoretical world where interest rates did not exist and monetary policy would work perfectly well. Imagining a IS/LM model where we have two assets. Money and equities. In such a world an increase in the money supply would push up the prices of equities. This would reduce the funding costs of companies and hence increase investments. At the same time it would increase holdholds wealth (if they hold equities in their portfolio) and this would increase private consumption. In this world monetary policy works perfectly well and the there is no problem with a “zero lower bound” on interest rates. Throw in the real estate market and a foreign exchange markets and then you have two more “channels” by which monetary policy works.
Hence, the Market Monetarist perspective on monetary policy the following dictum holds:
“Monetary policy works through many channels”
Keynesians are still obsessed about interest rates
Fast forward to the debate today. New Keynesians have mostly accepted that there are ways out of the liquidity trap and the work of for example Lars E. O. Svensson is key. However, when one reads New Keynesian research today one will realise that New Keynesians are as obsessed with interest rates as the key channel for the transmission of monetary policy as the old keynesians were. What has changed is that New Keynesians believe that we can get around the liquidity trap by playing around with expectations. Old Keynesians assumed that economic agents had backward looking or static expectations while New Keynesians assume rational expectations – hence, forward-looking expectations.
Hence, New Keynesians still see interest rates at being at the core of monetary policy making. This is as problematic as it was 30 years ago. Yes, it is fine that New Keynesian acknowledges that agents are forward-looking but it is highly problematic that they maintain the narrow focus on interest rates.
In the New Keynesian model monetary policy works by increasing inflation expectation that pushes down real interest rates, which spurs private consumption and investments. Market Monetarists certainly do think this is one of many channels by which monetary policy work, but it is clearly not the most important channel.
Rules are at the centre of the transmission mechanism
Market Monetarist stresses the importance of monetary policy rules and how that impacts agents expectations and hence the monetary transmission mechanism. Hence, we are more focused on the forward-looking nature or monetary policy than the “old” monetarists were. In that regard we are similar to the New Keynesians.
It exactly because of our acceptance of rational expectations that we are so obsessed about NGDP level targeting. Therefore when we discuss the monetary policy transmission mechanism it is key whether we are in world with no credible rule in place or whether we are in a world of a credible monetary policy rule. Below I will discussion both.
From no credibility to a credible NGDP level target
Lets assume that the economy is in “bad equilibrium”. For some reason money velocity has collapsed, which continues to put downward pressures on inflation and growth and therefore on NGDP. Then enters a new credible central bank governor and he announces the following:
“I will ensure that a “good equilibrium” is re-established. That means that I will ‘print’ whatever amount of money is needed so to make up for the drop in velocity we have seen. I will not stop the expansion of the money base before market participants again forecasts nominal GDP to have returned to it’s old trend path. Thereafter I will conduct monetary policy in such a fashion so NGDP is maintained on a 5% growth path.”
Lets assume that this new central bank governor is credible and market participants believe him. Lets call him Ben Volcker.
By issuing this statement the credible Ben Volcker will likely set in motion the following process:
1) Consumers who have been hoarding cash because they where expecting no and very slow growth in the nominal income will immediately reduce there holding of cash and increase private consumption.
2) Companies that have been hoarding cash will start investing – there is no reason to hoard cash when the economy will be growing again.
3) Banks will realise that there is no reason to continue aggressive deleveraging and they will expect much better returns on lending out money to companies and households. It certainly no longer will be paying off to put money into reserves with the central bank. Lending growth will accelerate as the “money multiplier” increases sharply.
4) Investors in the stock market knows that in the long run stock prices track nominal GDP so a promise of a sharp increase in NGDP will make stocks much more attractive. Furthermore, with a 5% path growth rule for NGDP investors will expect a much less volatile earnings and dividend flow from companies. That will reduce the “risk premium” on equities, which further will push up stock prices. With higher stock prices companies will invest more and consumers will consume more.
5) The promise of loser monetary policy also means that the supply of money will increase relative to the demand for money. This effectively will lead to a sharp sell-off in the country’s currency. This obviously will improve the competitiveness of the country and spark export growth.
These are five channels and I did not mention interest rates yet…and there is a reason for that. Interest rates will INCREASE and so will bond yields as market participant start to price in higher inflation in the transition period in which we go from a “bad equilibrium” to a “good equilibrium”.
Hence, there is no reason for the New Keynesian interest rate “fetish” – we got at least five other more powerful channels by which monetary policy works.
Monetary transmission mechanism with a credible NGDP level target
Ben Volcker has now with his announcement brought back the economy to a “good equilibrium”. In the process he might have needed initially to increase the money base to convince economic agents that he meant business. However, once credibility is established concerning the new NGDP level target rule Ben Volcker just needs to look serious and credible and then expectations and the market will take care of the rest.
Imagine the following situation. A positive shock increase the velocity of money and with a fixed money supply this pushed NGDP above it target path. What happens?
1) Consumers realise that Ben Volcker will tighten monetary policy and slow NGDP growth. With the expectation of lower income growth consumers tighten their belts and private consumption growth slows.
2) Investors also see NGDP growth slowing so they scale back investments.
3) With the outlook for slower growth in NGDP banks scale back their lending and increase their reserves.
4) Stock prices start to drop as expectations for earnings growth is scaled back (remember NGDP growth and earnings growth is strongly correlated). This slows private consumption growth and investment growth.
5) With expectations of a tightening of monetary conditions players in the currency market send the currency strong. This led to a worsening of the country’s competitiveness and to weaker export growth.
6) Interest rates and bond yields DROP on the expectations of tighter monetary policy.
All this happens without Ben Volcker doing anything with the money base. He is just sitting around repeating his dogma: “The central bank will control the money base in such a fashion that economic agents away expect NGDP to grow along the 5% path we already have announced.” By now he might as well been replaced by a computer…
…..
Recommended reading on the “old” monetarist transmission mechanism
Milton Friedman: “Milton Friedman’s Monetary Framework: A Debate with His Critics”
Karl Brunner and Allan Meltzer: “Money and the Economy: Issues in Monetary Analysis”
For a similar discussion to mine with special focus on the Paradox of Thrieft see the following posts from some of our Market Monetarist friends:
Josh Hendrickson
David Beckworth
Bill Woolsey
Nick Rowe
And finally from Scott Sumner on the differences between New Keynesian and Market Monetarist thinking.
—-
Update: Scott Sumner has a interesting comment on central banking “language” and “interest rates”.
Marcus Nunes
/ October 30, 2011Lars
Yes, the interest rate fetish. This is likely due to Bernanke´s “creditist views”. In his Money & Banking Text, Mishkin lists 10 channels. Recently I did a post where interest rates are not mentioned at all!
http://thefaintofheart.wordpress.com/2011/09/15/an-alternative-script-%E2%80%93-one-where-interest-rates-don%C2%B4t-make-an-%E2%80%9Cappearance%E2%80%9D/
Fmb
/ October 30, 2011I accept that this works over a wide range. But, what happens in extreme scenarios, where eg all liquidity is soaked up, or the central bank owns all assets? Could the threat of such scenarios lead to some unraveling of behavior in regions where we’d otherwise hope/expect things to work?
Budget constraints would interfere with attempts to test (the original) Volcker’s resolve to tame inflation, but no such constraint binds if we try to test Ben’s resolve.
Sorry if I’m not very clear.
Alex Salter
/ October 30, 2011Well stated. The interest rate isn’t the relevant channel; expectations and the money demand function are. If the public gets nothing else from Market Monetarism, I really hope they get this.
Lars Christensen
/ October 30, 2011Marcus, there is no need to spend too much time on the interest rate channel. That we certainly can agree on. I am, however, not sure that Bernanke is being fooled – I however believe that he is facing a “institutional liquidity trap”. So even if he want to do the right thing the institutional limits within the Federal Reserve as an institution is stopping him.
Fmb, that seems to be a rather extreme scenario. I guess that it possible, but I have a very hard time seeing that has empirically relevant. In fact I believe that for the US it would not take a lot of buying of assets to push NGDP back on track once expectations turn around.
In my view it is likely that the Fed has ended up far to many assets to get NGDP back on track. The problem with QE1 and QE2 was that these measures were pushed through without any target of any kind being announced. It has always seemed odd to me that the Fed said how many and what assets it would buy with out stating what it wanted to achieve.
Alex, thanks. I hope we will achieve a bit more, but yes, making people understand that monetary policy and interest rates has very little to do with each other is an extremely important task for us all.
Marcus Nunes
/ October 30, 2011Lars:”So even if he want to do the right thing the institutional limits within the Federal Reserve as an institution is stopping him”.
Just like “diffrent constraints” stopped Romer from giving good advce 18 months ago when she was still CEA head!
http://thefaintofheart.wordpress.com/2011/10/30/the-romer-%E2%80%9Cconversion%E2%80%9D-and-suggestion-of-a-volcker-moment/
Lars Christensen
/ October 30, 2011Marcus, I think sadly that you are right…and don’t forget Romer gave pretty bad advise on fiscal policy in my view.
Lorenzo from Oz
/ October 30, 2011Very helpful. Some typos: it should be Brunner-Meltzer critique of; and here the work of for example Lars E. O. Svensson is key; in the country’s currency; been replaced by a computer.
Benjamin Cole
/ October 30, 2011There was a time when President Carter “leaked” a quote, to the effect he “would kick Ted Kennedy’s ass” if Kennedy decided to run. It was an intentional leak.
While I believe in transparent, rules-based monetary policy, at this time Bernanke is seen as feeble. Jeez, even Romer is telling to act tough.
Bernanke needs to be “overheard” at the Fed cafeteria, telling a staffer, “We are going to print money until we are drowning in the stuff,” or something to that effect.
Lars Christensen
/ October 30, 2011Lorenzo, thanks – speed is key here, but I have now updated the text. Thanks for the input.
PS I am sure there must be at least ten more typos. The great thing with blogging is that it is exactly as speaking – it is nearly realtime.
Brito
/ October 31, 2011This all makes sense given your assumptions, but I still don’t understand why we should assume that investors & banks will actually believe that Volcker will raise NGDP to the desired level just because he says he will. Expectations alone cannot be the transmission mechanism, if investors have rational expectations then their expectation must be justified on a real mechanism that will inherently raise NGDP at some point in the future irrelevant of expectations. What is the mechanism?
Lars Christensen
/ October 31, 2011Brito, monetary policy is all about credibility. The question is how does the central bank build credibility. Real Volcker did that back in 1979 (or rather in 1982 if you ask Scott Sumner – and me for that matter) by acting. He announced money supply targets and tried to bring money supply growth by hike interest rates. That started to work in 1982 – at least judging from market movements (and the drop in inflation). I took 2-3 years before the markets really started to buy into the story. I would however claim that it is much easier to convince investors that you will increase NGDP (and inflation in the short-term) compared to the politically much more challenging situation Volcker was facing.
A simple rule could be the following. Ben Volcker announced that he want people to believe him so every Friday he will put X$ on the main square of a city randomly chosen somewhere in the country. He will doubly this amount until the market starts to price in the “right” expectations for the NGDP path. He could of course also just use the T-bills markets. Alternatively he could buy commodities – steel, gold, silver or whatever – it is pretty easy to get money in circulation – also with a banking system in crisis.
Dan Kervick
/ October 31, 2011“In such a world an increase in the money supply would push up the prices of equities.”
“I will ensure that a “good equilibrium” is re-established. That means that I will ‘print’ whatever amount of money is needed so to make up for the drop in velocity we have seen. I will not stop the expansion of the money base before market participants again forecasts nominal GDP to have returned to it’s old trend path. Thereafter I will conduct monetary policy in such a fashion so NGDP is maintained on a 5% growth path.”
Lars, let’s consider two simplified thought experiments, representing the extremes of money printing operations.
In the first, the Fed orders up $15 trillion in new Federal reserve notes. They are delivered to the Fed in boxes in container trucks. The boxes are offloaded, and packed away in a warehouse under lock and key, with orders from the central banker that they never be opened. Everyone in America knows this has happened. Everyone knows what the orders are.
In the second thought experiment, the central bank again orders up $15 trillion in new Federal Reserve notes. They are again delivered to the Fed in boxes in container trucks. But this time, Fed employees open the boxes and mail $50,000 to each living soul in the United States. Again, assume everyone in America knows this has happened. Everyone knows what the policy decision and actions are.
I think we can agree that in the first case, the money printing would have very little impact on prices, while in the second case, the money printing would have a profound impact on prices.
(Predicting the actual impacts however is not that easy in the real world, because they depend not on the idealizations of rational expectations theory, but on whatever wild and quirky theories regular people have about the nature of money and prices, theories which are sometimes bizarre or irrational.)
Now the Fed has several operational tools available to it that fall somewhere along the spectrum between the circumstances of the first thought experiment and the circumstances of the second thought experiment. And the response of people who pay attention to Fed moves surely depends on their behavioral responses not just to Fed statements about what its alleged targets are, but to more detailed beliefs about what exactly the Fed is going to do. Specifically, we can expect very different responses based on the degree to which observers believe Fed actions are more like the first thought experiment, or more like the second thought experiment. And these differences in response hold, even if we assume that in all cases the observers are firmly convinced of the Fed’s commitment to carry out their announced policy.
Personally, I think “opening the boxes” – figuratively speaking – and adding the money as balances to bank reserve accounts is a lot more like putting the boxes in a warehouse than sending money to people. So I personally wouldn’t expect to see many changes if I thought that was the Fed’s main available policy mechanism. But that’s just me. Other people might expect different results.
So while it’s good that you and the other MMers are beginning to address these questions about transmission mechanisms, so far the story you are telling still seems overly dependent on the power of the beliefs in the commitment, and not the details of the beliefs about the tools that will be used in pursuing the commitment.
It still looks to me like the transmission mechanism is just a black box in a flow diagram on which someone has written, ” … and then an amazing change in expectations happens.”
Brito
/ October 31, 2011Lars, I agree that getting money into circulation and growing it will increase NGDP. But I don’t think the Fed can do this easily, quantitative easing doesn’t really get money circulated amongst businesses and ordinary people, the cash remains motionless at banks’ reserves at the fed, it is not being lent out. So I can’t see a non interest rate transmission mechanism from QE to higher NGDP, so the central bank will seriously have to consider buying other things, perhaps from institutions other than banks, or even directly finance government expenditure. Unfortunately this is something Bernanke has basically ruled out as illegal or outside the roles of the Fed in his testimony to congress.
Lars Christensen
/ October 31, 2011Brito, what would happen is the Federal Reserve put a trillion dollars on time square every Friday for a month? Or the Bank of England bought a trillion euros every week. Or the Czech central bank bought gold for a trillion Czech koruna a week? Or the Riksbank paid for paid all Swedish publish expenditures for a year?
It would all be highly inflationary (and no, I do not recommend these central banks to test it….)
Brito
/ October 31, 2011Lars, I agree with that. I’m not saying the bank /theoretically/ can’t increase monetary circulation sufficiently. All I’m saying is that it appears to me unlikely that the bank will do anything other than lower IOR and do more QE, lowering IOR might help, but isn’t that essentially an interest rate mechanism and thus ruled out? QE on the other hand I just can’t see actually leading to money entering general circulation, rather than staying in excess reserves (but maybe I’m wrong?). Is there any action in particular you think the bank should do to increase money in circulation?
Christopher Mahoney
/ February 11, 2016I prefer NPLT. Simpler and easier to explain.