Archive for March 2019


There’s only one reliable rule of thumb in macro

This is a follow-up to the previous post.

There is an endless search for rules of thumb in macro.

In the 1940s, very low interest rates were the new norm.

In the 1950s, rates began rising and frequent mild recessions were the new norm.

In the 1960s, one long “Phillips Curve” expansion was the new norm. We had it all figured out.

In the 1970s, the Phillips Curve fell apart, and we just had to live with stagflation.

In the 1980s, we didn’t have to live with stagflation, but big deficits were the new norm.

In the 1990s, we achieved budget surpluses and a Great Moderation (noninflationary boom), something no one expected.

In the 2000s, the Great Moderation collapsed into a deep recession that few expected (certainly not me or Robert Lucas.) Also, America’s first big housing boom and bust. Also, bank runs that were supposedly ended by FDIC.

In the 2010s, we had near-zero interest rates even as the economy recovered and unemployment fell to moderate levels. Also unexpected.

Every decade produces a new and unexpected macro situation and the 2020s will be no different. Rules of thumb don’t hold up over time.

So don’t tell me, “When you look at history, it’s clear that X will happen.”

Sorry, but there’s only one reliable rule of thumb in macro:

Things change.

PS. I am reluctant to hazard a guess as to what will make the 2020s special; perhaps it will violate the rule of thumb that says, “American expansions never last more than 10 years.”

PPS. I have a post on the Steve Moore nomination at Econlog.

PPPS. But don’t read the Steve Moore post, read this one.

The yield curve inverted

The Financial Times reports that the yield curve inverted, albeit by just 1 basis point:

Uh oh

Does this mean a recession is more likely than before?


Does this mean that a recession is likely in the next 12 months?

Probably not.

Does this mean that monetary policy is too tight?

Hard to say.

In my view, the current situation reminds me most closely of 1998, when the US economy was pretty strong at a time when the global economy was weakening. That year also saw a mild inversion, which turned out to be an incorrect recession forecast:

Ah, more reassuring

In its entire history, the US has never experienced a soft landing, in the sense of having a recovery extend for many years after unemployment had fallen close to the natural rate. Since unemployment has fallen close to the natural rate, that fact alone should make us worried. On the other hand:

1.Australia and the UK have experienced soft landings in recent decades.

2. In previous US expansions, you didn’t have MMs beating the drum for stable NGDP growth, level targeting. (Yes, I’m joking.)

When unemployment falls to the natural rate, problems are almost always just around the corner. In 1966, the yield curve inverted when unemployment fell below 4%, and again there was no recession. But there was a problem just around the corner—high inflation.

So low unemployment is a very dangerous time for the US. We are a clumsy country, which fails to achieve soft landings. That’s the real story here, which is even deeper that the “inverted yield curve means recession” story. Low unemployment and inverted yield curves are a warning of choppy waters ahead, of monetary policy in danger of drifting off course, one way or another.

In my view, the stock market decline at the end of 2018 was partly due to a perception that rates should not be increased, while the Fed was planning another set of rate increases in 2019. (Other worries such as trade, global growth, etc., also played a role).

When the Fed backed off, stocks rallied. Despite today’s sell-off, stocks are at very lofty levels and hence stock market investors likely don’t foresee a recession during the next 12 months. Neither do I. But it would be foolish to deny that the recent yield curve inversion makes a recession more likely than before, say a 25% risk rather than a 15% risk.

Update: This recent post had more to say on the yield curve issue.

MMT “explained”

Bloomberg has a new article that attempts to explain what MMT is all about. It begins by quoting yours truly:

“MMT has constructed such a bizarre, illogical, convoluted way of thinking about macro that it’s almost impervious to attack,” Bentley University economist Scott Sumner claimed recently on his blog

Their explanation begins as follows:

A good place to start is with a simple description that you can carry in your pocket: MMT proposes that a country with its own currency, such as the U.S., doesn’t have to worry about accumulating too much debt because it can always print more money to pay interest. So the only constraint on spending is inflation, which can break out if the public and private sectors spend too much at the same time. As long as there are enough workers and equipment to meet growing demand without igniting inflation, the government can spend what it needs to maintain employment and achieve goals such as halting climate change.

If you’ve absorbed that much, you’re already ahead of a lot of the critics. 

Unfortunately, I’m not ahead of the critics, as I have no idea what that means. All I can ascertain is that they don’t think we need to worry about accumulating too much debt. I say that because they say we shouldn’t worry about accumulating too much debt. But is even that interpretation accurate? Not in the world of MMT:

Another misconception is that MMT says deficits never matter. On March 13 the University of Chicago Booth School of Business published a survey of prominent economists that misrepresented MMT that way, leaving out its understanding that too-big deficits can cause excessive inflation.

OK, so perhaps we do need to worry about accumulating too much debt, as it could lead to high inflation? I’m confused.

MMT also draws on the “functional finance” work of the Russian-born British economist Abba Lerner, who wrote in the 1940s that government should spend what’s required to achieve its goals, deficits be damned. 

OK, so now deficits don’t matter anymore. Whatever you accuse the MMTers of saying, they can find a quote that says the exact opposite. Now you might argue that they’re only saying deficits don’t matter when they are directed toward “goals”, and we all know that the goals of progressive politicians are a clearly delineated highly scientific concept. “We need precisely $X trillion to meet our goals, and not a penny more. Any additional spending beyond $X trillion would exceed our “goals” and lead to inflation.” Is that the idea?

The article is packed full of puzzling claims:

Modern Monetary Theory says the world still hasn’t come to terms with the death of the gold standard in 1971, when President Richard Nixon declared that the dollar was no longer convertible into gold. In the modern era of “fiat” currency, MMT says, the U.S. and other big economies no longer need to worry about having enough gold to back their paper money, so they’re free to print however much they need.

I think mainstream economists understand that money is not backed by gold and that the central bank can technically print almost unlimited amounts (barring government restrictions). The debate is over the effects of doing this.

Unless they mean that some economists don’t understand that central banks always have enough “ammo” to hit their target. But do even MMTers understand that?

MMT claims to be the legitimate heir to the theories of Britain’s John Maynard Keynes, who created the field of macroeconomics during the Great Depression. Keynes coined the term “paradox of thrift.” His insight was that while any single household can dig itself out of a hole by cutting spending when its income falls, the economy as a whole cannot. One household’s spending is another’s income, so if everybody cuts back, no one gets paid.

Of course this is wrong, as even modern Keynesians recognize. The economy can reduce overspending by shifting resources from consumption toward saving/investment. The final sentence confuses “consumption” with “spending”, an EC101 level error. At an aggregate level, spending includes investment spending. One can debate the zero bound problem, but this article seems to imply the paradox holds when not at the zero bound. it doesn’t.

MMT rejects the modern consensus that economies should be steered primarily by the raising and lowering of interest rates. MMTers believe that the natural rate of interest in a world of fiat money is zero and that pegging it higher is a giveaway to the investor class. They say tweaking interest rates is ineffectual because businesses make investment decisions based on prospects for growth, not the cost of money

The first and final sentences are defensible, but what comes between . . . oh boy! First of all, there is no distinction made between real and nominal interest rates. I have to believe they mean real interest rates, as the alternative to is too horrible to contemplate. But why are natural real interest rates zero, and how does the central bank “peg” real interest rates?

MMT says that, contrary to appearances, banks don’t make loans out of deposits. Rather, they make loans based on the demand for borrowing, then the borrowers stash the proceeds in the bank. Anyone they write a check to simply makes a deposit in another bank. The bottom line is that loans create deposits rather than deposits creating loans.

It may not make sense to talk about deposits creating loans, but it’s equally silly to talk about loans creating deposits. As Krugman once said, “it’s a simultaneous system.”

MMT challenges a core principle of conventional economics, which is that an increase in budget deficits will tend to raise interest rates, all else equal. Just the opposite, it says, sounding a bit like the White Queen from Alice in Wonderland. When the government spends more, the private sector gets the money and puts it in the banking system. With more money in the system and no increase in demand for it, interest rates will tend to fall, not rise, MMT says. That is, unless the government chooses to soak up reserves by selling bonds, which it doesn’t have to do.

So economists argue that debt financed budget deficits raise interest rates, and MMTers respond that they are wrong if you assume the deficit is not debt financed. Okaaay.

And what is the level of nominal interest rates in Latin American countries that relied on money creation to finance their deficits, not debt? I know, they didn’t follow some other provision of MMT. Stephanie Kelton once cited Japan as providing an example of how to achieve low rates with big deficits. Yes, you can hold down rates with near-zero NGDP growth over 25 years—that will “work”.

MMTers hold that inflation isn’t primarily the result of excessively strong growth. They blame much of it on businesses’ excessive pricing power. So before trying to choke off growth to kill inflation, they would try to break up monopolies and stop banks from making too many loans. 

This theory was discredited by the early 1980s, and for good reason. It doesn’t explain either time series or cross sectional variation in inflation rates.

Critics of MMT reject its reassurance that a country with its own currency doesn’t need to worry about deficits. After all, it’s been proven that a nation that loses the confidence of the world’s investors will see its currency plummet. As recently as 1976, the U.K. was forced to appeal to the International Monetary Fund to stabilize the value of sterling. Wray said the U.K.’s mistake was trying to peg its currency to the dollar and the crisis eased when it allowed the pound to float.

Devaluation? What could go wrong? Have you checking the UK’s inflation rate during the 1970s?

MMT’s critics argue that trying to use fiscal policy to steer the economy is a proven failure because Congress and the president rarely act quickly enough to respond to a downturn. And they say politicians can’t be relied upon to impose pain on the public through higher taxes or lower spending to squelch rising inflation. MMTers respond that they also oppose fine-tuning and instead want to use automatic stabilizers—including the jobs guarantee—to keep the economy on track.

So we are going to achieve our 2% inflation target with “automatic stabilizers”? And what happens when President Trump decides to massively increase the budget deficit when the economy doesn’t need more spending? Who will keep inflation on target, if not the Fed?

What’s more surprising is how much flak the school of thought is taking from liberal economists who’d appear to be natural allies, such as Larry Summers, the former Treasury secretary and former Harvard president.

This might be the most puzzling comment of all. You tell all these brilliant Ivy League macroeconomists that they are fundamentally wrong in their understanding of macroeconomics, and that only a few people at places like UM Kansas City understand what’s really going on, and you are surprised that they are skeptical? Say what you will about my views; I’m not surprised that MM was not welcomed with open arms.

I sometimes have commenters discussing whether MMT “works”. That’s not even a question. Something can’t “work” unless it’s a coherent policy that can be explained and then adopted. MMT has not even reached the stage where it’s possible to discuss whether it “works”, or whether the data tends to confirm its views. For that, they need a hypothesis that impartial observers can understand.

Please don’t ever leave another comment discussing whether MMT “works”, or I might ask you to explain what it is. And I am 100% sure you will fail, because I can always find a quote from this article that proves your interpretation is wrong.

HT: Dilip

Whatever floats your boat

[If you only have time for one post today, my Econlog post is the better choice.]

Some people seem to feel a deep need to announce one “bubble” after another. Here’s the FT announcing another American “housing bubble”. I guess as long as this practice contributes to mental health, I have no objection. Just don’t let the belief in bubbles alter your behavior in any way.

Here’s The Economist, in 2011:

The bursting of the Bitcoin bubble

BITCOIN, briefly the world’s favorite cryptocurrency, is in trouble. It plummeted from a peak of around $33 per unit in June to just $2.51.

Whew! So glad I didn’t buy Bitcoin at $33. Thanks Economist.

And here’s CNN Business in 2013:

The price of Bitcoins has plunged more than 70% in the past two days, sparking a rush of activity that overwhelmed trading platforms and suggested the bubble in the virtual currency has burst. Bitcoins were down to $77.56 as of 3 p.m. ET Friday.

Prices reached as high as $266 per Bitcoin around 7:30 a.m. ET Wednesday. But the price started to fall through the rest of day and Thursday morning.

And here’s CNN Business in 2015:

Bitcoin lost more than 60% of its value last year. The digital currency has already plunged another 30% in the first few days of 2015 — and that includes a 30% rebound on Thursday!

The price of Bitcoin (XBT) briefly fell below $200 on Wednesday, an important psychological barrier, before bouncing back. . . .

Bitcoin was a bubble that has burst. Jeffrey Gundlach, head of influential investment firm DoubleLine, is firmly in the Bitcoin bear camp. In a webcast on Tuesday, Gundlach declared that Bitcoin is “on its way to being relegated to the ash heap of digital currencies.”

Now in 2019, Bitcoin prices are up to nearly $4000 and there is even more bubble talk:

We can safely say the ICO bubble is over now.

“Safely”? We can now have confidence on where Bitcoin prices will be next year? Next decade? Next century?


PS. You say, “bubbles are hard to define but I know one when I see it”. Really? How about an SUV? Do you know an SUV when you see one? Below I provide two pictures of cars that are listed as SUVs:

An SUV when I was young
An SUV today

HT: Tyler Cowen

What role do we want the monetary base to serve?

Over at Econlog, I have a new post discussing the Fed’s opposition to narrow banking, and specifically John Cochrane’s excellent post criticizing the Fed’s position. I’ll eventually get to narrow banking in this post, but first I’d like to consider some basic questions about the monetary base, which are rarely asked.

Before 1913, the US had no Fed and the monetary base was 100% composed of currency (and coins.) There were no bank deposits at the Fed. It’s perfectly possible to run the world’s largest economy on that basis. Yes, there were occasional financial crises, but the 1931-33 banking crisis was far worse, so creating the Fed didn’t solve the problem. When I was in school, I was taught that it was FDIC that actually ended banking crises, but then we had 2008. Crisis-free Canada didn’t have a central bank until the 1930s and had no deposit insurance until the 1960s, so whatever causes financial crises it’s certainly not a lack of government intervention.

When the Fed was created in 1913, base money was expanded to include deposits at the Fed, not just currency. But why? Why add these reserve deposits, and if it’s a good idea, why not let anyone have a deposit at the Fed? I’ve never seen a good explanation.

I once wrote a post suggesting that we go back to the pre-1913 currency-only monetary base, half jokingly and half seriously. I’m sure John Cochrane would have been horrified, as he has quite rightly noted that the government can produce safe liquid assets quite cheaply, and hence on efficiency grounds there’s something to be said for saturating the economy with the stuff (say via the “Friedman rule”.)

But I wonder if that isn’t penny wise and pound foolish. Or as Tobin once said, “it takes a heap of Harberger triangles to fill an Okun Gap”. (Which means micro efficiency costs are an order of magnitude smaller than the costs of demand side recessions, i.e. the cost of inadequate monetary policy.)

Suppose we had had a currency-only monetary base in 2008. If the Fed had done $3 trillion in open market purchases, what would the impact of all that extra currency have been, given the currency stock was only $850 billion going into the recession? I’m not sure, but I suspect the Fed would have gotten more bang for the 3 trillion bucks.

Since my currency only idea is a pipe dream, let’s go the other direction. Let’s say deposits at the Fed are a great idea. I still want to know why they are restricted to banks. It’s only a matter of time before we go to all electronic money. Is it feasible to have a monetary regime where it is illegal for anyone but a banker to hold lawful money? I doubt it. In that case, it seems like Morgan Ricks’ proposal for allowing the public to have accounts at the Fed is only a matter of time.

Indeed the logic of this is so powerful that “narrow banks” are already trying to create a backdoor into the Fed, by having bank deposits where 100% of the money is placed in interest-bearing reserves. Basically, you’d have a checking account at the Fed, but there’d be a middleman to preserve a fig leaf for the idea that in the era of FDIC and interest-bearing reserve accounts, the commercial banking system is actually a useful way of providing liquidity to the public. It isn’t. Banks should be taking illiquid funds (CDs, etc.) and creating illiquid assets (loans, etc.).

Somewhere between ultra-safe reserves and risky loans we have bills and bonds. Who should serve as the intermediary in that case? Mutual funds?

If you want to have low inflation and zero interest rates, the central bank balance sheet will get large. Eventually the Fed may go beyond Treasuries and start buying private bonds. But big central bank balance sheets are a choice; it’s not hardwired into the modern world, as so many pundits seem to believe.

If you allow narrow banking, it may have the effect of increasing the demand for base money. John Cochrane suggests that the Fed has the option of not increasing the supply, which is true. But if rates are stuck at zero when demand for reserves increases, then not increasing the supply is highly contractionary.

So now we have another issue to consider, balance sheet size and monetary policy effectiveness. In theory, the government can create reserves costlessly, and meet any market demand. But real world central banks seem reluctant to have excessively large balance sheets, largely for psychological reasons. Unless this mental block can be overcome, perhaps through therapy, a greatly expanded central bank balance sheet might complicate monetary policy.

To me, central banks seem like a spoiled 10-year old boy, who complains about every option you offer him:

1. Do monetary policy “a outrance” to prevent demand shortfalls? They whine that there are “costs and risks” of doing what it takes.

2. But the costs and risks of the Fed buying government bonds are not real, as a central bank is part of the consolidated government balance sheet. Then they whine that it would be embarrassing to ask Congress for a “bailout”.

3. OK, adopt the sort of monetary regime that prevents the zero bound, one of the options now favored by many academic economists (as well as economists like Bernanke before he got to the Fed.) The Australian approach, for instance. Can’t do that, it’s too controversial.

So we’ve got this ever expanding Rube Goldberg device of a financial system, where plugging one leak just causes another to develop. Each step seems logical, but no one is thinking about the ultimate destination. What do we want of our monetary base? What types of base money should exist? Who should get to hold each type of base money? How attractive should it be to hold base money? (I.e. how competitive should the IOR rate be?)

The Fed’s horribly confused defense of its anti-narrow banking policy is an indication that there’s no real long run plan here; they are making it up as they go along. As each step exposes more logical contradictions, more patches are put in place, more plugs placed in leaks at the bottom of the boat. Our political system is not one where you can simply “blow it all up” and start with a more rational system, and maybe that’s for the best (for Hayekian reasons.) So we’ll keep muddling along, and eventually end up in a very different place.