That’s a made up quote, and perhaps a bit of hyperbole.  But it illustrates something that really bugs me about the blogosphere.  People make all sorts of claims about monetary policy near the zero bound:

1.  QE has no effect

2.  Negative interest rates would be contractionary

3.  Central banks cannot depreciate their currencies at the zero bound

And then when the theories are actually tested and the results are in, they keep making the same claims, seemingly oblivious to the fact that their theories have been discredited.

Today I’d like to talk about negative interest rates on reserves (IOR).  There are two broad approaches to this topic.  There’s what I’d call the “finance approach”, which claims negative IOR would actually be contractionary, because . . . well I’m not quite sure why. Something about how it interacts with the banking system.  I used to see these articles in the Financial Times.

I tell people to ignore banking when studying monetary policy—focus on the supply and demand for the medium of account (base money).  Negative IOR causes a fall in the demand for base money, and thus is expansionary.  Period.  End of story.

But just to make sure, let’s look at reality.  The Telegraph has a nice story on negative rates on Sweden, which was the first country to adopt the policy I proposed in early 2009.  (I deserve zero credit for NGDP targeting (the idea’s been around forever), but surely at least a tiny bit of credit for negative IOR.)  Before getting to the Telegraph story, let’s look at how markets responded to a recent negative IOR shock:

Updated March 18, 2015 11:14 a.m. ET

STOCKHOLM””Sweden’s central bank has slashed its key policy rate deeper into negative territory and expanded its bond-buying program to prevent the recent appreciation of the Swedish krona from stifling a budding revival in inflation.

The Riksbank, the Swedish monetary authority, lowered its benchmark rate to minus 0.25% from minus 0.1% and said it would buy government bonds worth 30 billion Swedish kronor ($3.45 billion), an extension of bond purchases worth 10 billion kronor announced earlier. The repurchase rate had stood at minus 0.1% since February, the first time it was cut into negative territory.

.  .  .

The Swedish krona fell sharply against the euro, which gained about 1% against the krona in the minutes after the announcement, hitting a high of over 9.34 kroner. Sweden equity markets jumped to a record high with the OMX Stockholm 30 Index up 1.5% at 1,700.

Yup, cutting the IOR is expansionary, even when in negative territory.  And those sorts of market responses (assuming at least partly unexpected, as some of them are) should have been it for the “finance” theories that negative IOR is contractionary, but you can never drive a stake through these zombie theories. There are still those “models” . . .

The Telegraph story looked at some macro data, which at first glance looks like a mixed bag.  Inflation has been running around negative 0.1% to minus 0.2% for 4 years:

It looks like 3 years, but that’s a quirk of year over year data, the actual CPI shows it’s 4 years:

So that doesn’t look very good for negative IOR.  But recall that in most countries inflation has recently been falling sharply, due to the plunge in oil prices.  Inflation has never been a reliable indicator.  NGDP growth has recently been accelerating in Sweden (up at a 9% annual rate in 2015, Q2), and the most recent RGDP growth shows 3.2% over the past year, which is higher than in previous years.  The best evidence comes from the unemployment rate, which leveled off at 8% after the Riksbank’s disastrous monetary tightening of 2011, but has recently fallen to 7%.

One final graph, which is really weird.  While other countries are seeing a surge in currency hoarding, Sweden is experiencing exactly the opposite, despite the negative rates. What’s up in Sweden?  Will the Nordic countries be the first to adopt the sort of cashless 1984-style panopticon state that is so beloved by authoritarians and Puritans everywhere?

PS.  I have a reply to John Cochrane over at Econlog.

Picture a crime movie where the killer is arrested.  After hours of being grilled under a hot incandescent bulb, he breaks down and confesses, sobbing as he describes how he murdered his wife so he could get the insurance money and run off with a another woman.  It would be very odd if the killer calmly confessed to that crime, with the demeanor of someone describing a walk in the park, or taking the kids out to fly a kite. You expect remorse. Indeed a calm demeanor might be taken as evidence of insanity; maybe the killer didn’t even understand what he had done.

Let’s set the stage.  It’s early October 2008.  It’s been three weeks since Lehman failed, and the global economy is starting to fall apart.  There are scary stories about trade plunging all over the world.  The stock market is crashing, falling continuously during the first 10 days of October, often by large amounts.  TIPS spreads are plunging. Commodity prices are plunging.  Commercial real estate is beginning to fall sharply, after surviving the initial subprime crisis.  Unemployment is rising sharply.  What does the Fed decide to do?  Commenter Beefcake sent me to the San Francisco Fed, for its explanation.

Before the crisis, Congress passed the Financial Services Regulatory Relief Act of 2006 authorizing the Federal Reserve to begin paying interest on reserves held against certain types of deposit liabilities. The legislation was supposed to go into effect beginning October 1, 2011. However, during the financial crisis, the effective date was moved up by three years through the Emergency Economic Stabilization Act of 2008.4 This was important for monetary policy because the Federal Reserve’s various liquidity facilities5 initiated during the financial crisis caused upward pressure on excess reserves and placed downward pressure on the Federal funds rate. To counteract these pressures, on October 6, 2008, the Federal Reserve Board announced that it would begin paying interest on depository institutions’ reserve balances.

Well there you are, what could be more sensible?  I’m impressed at the audacity of the Fed, if nothing else.  This is the explanation they provide in their “education” department.  I suppose the average man on the street would nod his head, thinking, “It sure looks like the Fed knew what they were doing.”

There’s just one problem, the Fed is describing a murder, the killing of the economy. Don’t be fooled by the calm tone of this explanation.  They are describing a policy that they themselves indicate was rushed into service three years early in order to prevent interest rates from falling.  Its entire purpose was to prevent monetary stimulus, to make money tighter.

Their explanation is that the money that was being put into the banking system to rescue Wall Street threatened to also reduce interest rates thus easing monetary policy. But that could have the side effect of also rescuing Main Street!  The Fed was so horrified by the thought of a cut in interest rates at the time the economy was falling off a cliff that they begged Congress to let them move earlier on IOR.

You might wonder why the Fed simply didn’t wave its magic wand, and set rates where they wanted them.  Actually, before they had the legal option of doing IOR, the Fed had no magic wand. It only seemed like they could directly control interest rates.  In fact, they controlled the monetary base, and used changes in the base to influence interest rates.  When they sharply boosted the base to rescue the banks in September 2008, they had no way of stopping interest rates from falling all the way to zero.

The week this happened I went absolutely ballistic, and essentially became a different person.  (After not have a strong opinion on monetary policy for the previous 25 years) I went down to Harvard to ask the top macroeconomists what the hell the Fed was doing, something I never would have done before 2008.

Now inevitably someone will cite Paul Krugman, who likes to point out that central banks are quite capable of driving the economy into a zero rate trap without IOR. The Japanese did it in the 1990s, and the Fed did it in the 1930s. That’s true but irrelevant. The fact that people can also be murdered with knives does not prove that a man standing over a dead body with a smoking revolver in his hand is not guilty of murder.

Others will say that one decision is no big deal; it’s the future path of policy that matters.  That’s also true, but decisions can tell us a lot about the Fed’s mindset, and hence the likely future path of policy.  As Tim Duy recently pointed out, the reason the markets care so much about the measly quarter point increase is not the direct effects on the economy, but rather what it tells us about the Fed’s reaction function.

The October 8, 2008 rise in IOR was neither a necessary nor sufficient condition to produce the Great Recession.  But it was one important part of the story, which began in 2008 and still has not ended.  Believe it or not a rate rise in September 2015 would help cause the recession of 2008, as the severity of that recession was partly caused by the perception that the Fed would raise rates before we got back to the previous NGDP trend line.  (An idea Krugman developed back in 1998.) And now the Fed is about to confirm that perception.

PS.  Now reread the final sentence of the opening paragraph.

PPS.  I have a new post at Econlog on a related issue.

When I advocate something like QE or negative interest on reserves, I often get people complaining that this will not boost bank lending, or that we shouldn’t even be trying to boost bank lending.  It almost makes me want to tear out my hair. What in the world does banking have to do with monetary policy?  Yes, it may or may not boost bank lending, but it doesn’t matter, as monetary policy is about the hot potato effect.  And yes, the Fed should not be trying to boost lending, any more than it should try to boost sales of microwave ovens.  NGDP is what matters.

Jim Glass directed me to a new study by the Bank of England, which confirms that monetary policy is about the hot potato effect (aka portfolio rebalancing channels) not bank lending.  Here is the abstract.

We test whether quantitative easing (QE) provided a boost to bank lending in the United Kingdom, in addition to the effects on asset prices, demand and inflation focused on in most other studies. Using a data set available to researchers at the Bank, we use two alternative approaches to identify the effects of variation in deposits on individual banks’ balance sheets and test whether this variation in deposits boosted lending. We find no evidence to suggest that QE operated via a traditional bank lending channel (BLC) in the spirit of the model due to Kashyap and Stein. We show in a simple BLC framework that if QE gives rise to deposits that are likely to be short-lived in a given bank (‘flighty’ deposits), then the traditional BLC is diminished. Our analysis suggests that QE operating through a portfolio rebalancing channel gave rise to such flighty deposits and that this is a potential reason that we find no evidence of a BLC. Our evidence is consistent with other studies which suggest that QE boosted aggregate demand and inflation via portfolio rebalancing channels.

PS.  I have a new post over at Econlog that is far more important than this post.  Read the whole thing.

My very first blog post after the intro discussed negative IOR.  I also published a couple short articles discussing the option in early 2009.  Six years later, what can we say?

1.  I was wrong in assuming the zero lower bound was only 1 or 2 basis points negative.  I made that assumption because in the 1930s and early 1940s T-bill yields never went more than a couple basis points negative.

2.  Being wrong about the zero lower bound made me even righter than I anticipated about the effectiveness of negative IOR.  I argued that people and institutions didn’t particularly want to hold huge amounts of currency, and that assumption was much truer than even I expected.

3.  There was a period where some contrarians were suggesting that negative IOR is a contractionary policy.  I thought that was wrong, as it reduces the demand for the medium of account.  Now I think it’s pretty clear that the contrarian view is wrong.  Negative IOR weakens a currency in the forex market.

4.  Points 1, 2 and 3 seem to make negative IOR a more attractive policy, but if anything my views have evolved in the other direction.  I’d prefer to focus on monetary policy tools like QE and forward guidance, which allow you to exit the zero bound more quickly.  Ultra low rates mean policy has been too tight.

5.  When I originally proposed the idea it was seen as being slightly wacky (even by me.)  Now you have negative yields on 8-year German bonds and 10-year Swiss bonds.  Negative rates are an important part of the modern financial world. Lesson?  Never say never.  NGDP futures might look like a wacky idea today, but back in 2009 the idea of negative yields on 10-year government bonds seemed far, far wackier.  We always need to search for the best options, and let the conventional wisdom catch up when it’s ready.

PS.  Back in early 2009, MMs were the only people saying monetary policy was MUCH too tight.  Matt Yglesias points out that recent events suggest that we were right:

For the USA, the main implication [of negative interest rates] is that back in 2009 and 2010 the Federal Reserve made a mistake. All the objective economic metrics at the time said the “right” interest rate to curb unemployment would be negative. But negative interest rates are impossible! The Fed tried a few tricks to get around that problem, and also told Congress to try fiscal stimulus as a workaround.

The implication of the European experience, however, is that the Fed could have generated negative interest rates through a mix of Quantitative Easing and negative interest rates.

Nick Rowe has a new post that explains what’s special about central “banks:”

What makes a central bank special?

The Bank of Canada can borrow and lend. So can the Bank of Montreal. So can I. Nothing special there.

The Bank of Canada can set any rate of interest it likes when it lends. So can the Bank of Montreal. So can I. Nothing special there. “If you want to borrow from me, you have to pay x% interest.” We can all say that. (Whether anyone will want to borrow from us at x% interest is another question.)

The Bank of Canada can set any rate of interest it likes when it borrows. So can the Bank of Montreal. So can I. Nothing special there. “If you want to lend to me, you have to accept y% interest.” We can all say that. (Whether anyone will want to lend to us at y% interest is another question.)

And yet there’s this utterly bizarre belief among many economists that it is the Bank of Canada that has the power to set Canadian interest rates, just by borrowing and lending. And that the Bank of Montreal, and ordinary people like me, somehow lack this special power. Even though we can borrow and lend too.

Now it’s true that the Bank of Canada is a lot richer than me. But what if I were a Canadian Bill Gates or Warren Buffet? And is the Bank of Canada richer than the Bank of Montreal? That can’t be the difference.

Now it’s true that the Bank of Canada can create money at the stroke of a pen, and I can’t. But the Bank of Montreal can. What makes the Bank of Canada special, compared to the other banks? What power does the Bank of Canada have that the Bank of Montreal lacks?

I’m going to give the same answer I gave nearly three years ago. And then I’m going to expand on it.

The fundamental difference between the Bank of Canada and the Bank of Montreal is asymmetric redeemability. The Bank of Montreal promises to redeem its monetary liabilities in Bank of Canada monetary liabilities. The Bank of Canada does not promise to redeem its monetary liabilities in Bank of Montreal monetary liabilities.

And then he ends the post as follows:

There is something very seriously wrong with any approach to monetary theory which says we can assume central banks set interest rates and ignore currency. It is precisely those irredeemable monetary liabilities of the central bank (whether they take the physical form of paper, coin, electrons, does not matter) that give central banks their special power. That’s what makes central banks central.

Now let’s consider a different monetary system.  Imagine a gold standard and a monopoly producer of gold.  The gold mine company would reduce short term interest rates by increasing the supply of gold.  Like currency, gold is irredeemable.  But no one would call this gold mining company a “bank” because it possesses no bank-like qualities.  Banks don’t create irredeemable assets, gold mines do.

Central banks may have some bank-like qualities, but what makes them special is their ability to produce currency–i.e. paper gold.  And that has no relationship to banking at all.

For years I’ve dealt with commenters who wanted to turn the discussion to banking:

“How do you know negative IOR will increase lending?”  I don’t care if it does, because banking has nothing to do with monetary policy.

“The Fed can’t cut rates any lower–how are they supposed to boost the economy?”  It doesn’t matter whether they can cut rates, because rates aren’t the transmission mechanism.

The Fed affects NGDP by changing the current supply and demand for the medium of account, and also the expected future path of the supply and demand.

“Monetary policy is already quite easy.”  No; credit is easy, monetary policy is ultra-tight.

In the comment section Nick says:

I just remembered. Back on the I=S post, IIRC, some people were complaining I didn’t talk about banks enough. OK, here’s a post on banks.

I’d say it’s a post on why central banks aren’t banks.