ABCs of Business Cycle Theory, Part 3
Hear Scott Sumner on why he prefers NGDP to inflation targeting, how the Federal Reserve missed the mark so badly in 2007–2008, the problem with a Gold Standard, and why cryptocurrency is no threat (yet) to central banking:
Part 1 — A Libertarian Visits the Federal Reserve, with Robert Wenzel
Part 2 — A Synthesis of Hayek, Minsky, Friedman and Keynes, with Brad DeLong
Or read the transcript of his interview with Charlie Deist:
Part 3
“ What Would Milton Say?” with Scott Sumner
Charlie Deist: Today I have the pleasure of speaking with Professor Scott Sumner. Scott Sumner is the Director of the Program on Monetary Policy at the Mercatus Center. He is also the author of The Midas Paradox: A New look at the Great Depression and Economic Instability, and he blogs at The Money Illusion.
Scott Sumner is not a household name, but behind the scenes he may be one of the most influential thinkers guiding what the economics profession thinks about the idea of nominal GDP targeting.
I was recently riding on the BART train here in the Bay Area when I noticed that the fellow standing next to me was reading some kind of economics paper. I saw over his shoulder that it was related to topic for this show. So I tapped this guy on the shoulder and asked him, “By any chance, have you heard of Scott Sumner?” I was going to tell him that I was reading his book, The Midas Paradox, and interviewing its author that weekend.
But he kind of shrugged me off, saying, “Oh, do you mean Larry Summers?” (Summers was a former Treasury Secretary who has a lot of influence).
I said, “No, no, Scott Sumner.”
The conversation didn’t go far, but he was kind enough to hand me the paper he was reading, because he was finished with it. The authors, Larry Summers and Olivier Blanchard, were looking at why we’ve had a persistent slow growth since the recession of 2007 to 2008 — what’s known as the Great Recession — and they gave a nod to this idea of nominal GDP targeting as one way to potentially improve the performance of monetary policy.
In reading it, I was struck by the fact that the people like Summers, who have the ear of policymakers, still tend to dismiss the idea that monetary policy was too tight and could have been much more effective in bringing the economy out of the recession. That is something Scott has been arguing convincingly on his blog.
The Silent Revolution: The Shift to Nominal GDP Targeting
Charlie Deist: Scott, when did you begin first begin your blog, The Money Illusion?
Scott Sumner: Well, first off, thank you for inviting me, Charlie. It started in February 2009. I was spurred to create the blog by my frustration with what I thought were missed opportunities in policymaking.
Charlie Deist: This was a couple of years after the housing crisis. You had a contrarian argument that monetary policy was too tight during the key period of the crash, and that the Federal Reserve could have done more. But what do these terms tight and loose mean in the context of monetary policy?
Scott Sumner: That gets to the heart of my disagreement with the conventional wisdom. Many people look at things like a change in the money supply or, even more so, interest rates to judge whether money is easy or tight, but those are actually not very good variables to focus on. It’s commonly thought that low interest rates are easy money and high interest rates are tight money, but just to give you one easy counterexample, if you have really easy money, which creates a lot of inflation, interest rates are actually at historic highs. We saw that in the 1970s for instance.
When you have really tight money, it sometimes leads to deflation or falling prices, and in that case interest rates often fall to zero. So, in fact, it’s not true that high interest rates are tight money, and low interest rates are easy money. It’s actually much more complicated than that, but that’s what threw people off track in 2008 and caused them to misdiagnose what was going on.
The Fed Misses the Mark
Charlie Deist: What kind of indicators were you looking at that clued you in to the fact that money might be getting tight in that period?
Scott Sumner: I looked at an indicator that was recommended by Ben Bernanke when he was an academic, and that is nominal GDP growth.
Nominal GDP is the total dollar value of the economy. For the average person, maybe the best way to think of it is if you added up everybody’s income in the economy, that would be close to nominal GDP.
The more common figure cited in the news media is real GDP, which strips out the effects of inflation, but it’s nominal GDP, or dollar spending, that best measures monetary policy. As a general rule, it should be running around 4–5 percent growth each year. If it’s very high, you’ll get very high inflation as a result. And if it’s too low, like in 2008 and 2009, you’ll get a deep recession.
Charlie Deist: Correct me if I’m off base here, but the standard thinking is that the Fed lowers the interest rate through open market operations — they buy and sell treasury bonds to adjust the rate at which banks lend to one another. That can in turn affect how likely banks are to loan out at certain interest rates. So, the Federal Reserve is said to be able to manipulate these short-term interest rates, but can it also affect longer-term interest rates?
Scott Sumner: Well, that’s exactly the problem. There are a lot of breaks in the links. Now, it is true that on a day-to-day basis they target the Federal funds rate. However, their ability to control that is very much limited by the conditions of the economy.
I’ll give you a recent example. Your listeners might recall that the Fed was telling people at the beginning of 2016, “We think we’ll raise interest rates maybe three or four times this year” — quarter point increases. In fact, they only raised interest rates one time, and the reason for that was very simple. The Fed really can’t push interest rates wherever it wants without the economy becoming very destabilized. So, to a very great extent, the Fed has to follow the economy in setting interest rates. While it’s true that they create a short-term target and vote on it in their meetings, they’re following trends in the economy to a much greater extent than people understand.
You really have to look at how monetary policy shapes the trends in the economy itself. In other words, let’s take a simple variable like inflation. To some extent, interest rates tend to follow inflation. They rise when inflation rises, so you have to look at how monetary policy affects the rate of inflation. If you print a lot of money, you can get high inflation under certain conditions, and that will lead to higher interest rates. In that case you, could actually have an easy money policy — i.e., printing a lot of money — creating high inflation and high interest rates.
If I’m right that interest rates mostly follow the economy and the Fed has to play along with where the economy is going, then the interest rate is not the right variable to see what the Fed is really doing. You want to look at the underlying trends in spending and the economy, and again, I think nominal GDP is probably the best single variable for doing that.
Charlie Deist: Central banks in most industrial countries have had some sort of inflation target of around 2 percent for several decades. Given that the Federal Reserve should have been looking at that indicator, what caused them to miss that target? We had a period of deflation briefly in 2008 and we’ve had persistently lower inflation ever since. There’s even some doubt that the Federal Reserve can even hit its inflation target, but you argue that the Federal Reserve should be able to hit whatever inflation target it wants. Where does the disagreement come from there?
Scott Sumner: There’s actually a couple questions embedded there. One is “Can the Fed hit any particular target?” — whether it be inflation or nominal GDP growth — and two is, “Which target is best?” Should it be inflation, or nominal GDP growth as I prefer?
Let’s just start with the second question. The reason I prefer a nominal GDP growth target is that inflation can reflect either demand side factors — that is, too much spending pushing up prices — or supply side factors — like a lack of oil in the global market leading to higher oil prices. Those actually have very different effects on the economy.
In the middle of 2008, we had a big spike in oil prices that pushed inflation up. For that reason, the Fed decided not to adopt a more expansionary monetary policy even though the housing and banking situations were already going downhill. In retrospect, the Fed clearly should have eased policy more during that year. Ben Bernanke even admitted that in his memoir, The Courage to Act. But they were worried about inflation
“The full FOMC would discuss NGDP targeting at its November 2011 meeting. We considered the theoretical benefits of the approach, but also whether it was desirable, or even feasible, to switch to a new framework at a time of great economic uncertainty. After a lengthy discussion, the Committee firmly rejected the idea. I had been intrigued by the approach at first but came to share my colleagues’ reservations about introducing it at that time. Nominal GDP targeting is complicated and would be very difficult to communicate to the public (as well as to Congress, which would have to be consulted). Even if we did successfully explain it, other challenges remained.
For nominal GDP targeting to work, it had to be credible. That is, people would have to be convinced that the Fed, after spending most of the 1980s and 1990s trying to quash inflation, had suddenly decided it was willing to tolerate higher inflation, possibly for many years. They’d have to be convinced that future Fed policymakers would continue the strategy, and that Congress would not act to block it.” [pp. 515–519, The Courage to Act, by Ben Bernanke]
If they had been looking at nominal GDP growth, they would have seen that actual spending was slowing during 2008, even as inflation was rising due to skyrocketing gasoline prices. Nominal GDP was giving a better indicator of the true nature of the economy. Looking back, we now know that 2008 was the beginning of a really severe recession, and a collapse in spending and demand in the economy. But the inflation number was hiding that fact, and if we had focused instead on nominal GDP targeting, the Fed would have been more expansionary in 2008 and the recession would have been milder.
The second question you raised is a good one. Can they hit any of those targets? The second point I would make is this: even if we have inflation targeting, the Fed did not do an adequate job because even inflation started falling at the end of 2008 and into 2009. As you point out, since 2008, the Fed has generally undershot its two percent inflation target, and that has led some people to wonder, “Well, are they even capable of creating two percent inflation?” That question, I think, has a very clear answer:
A central bank that has what economists call fiat money, which people usually call paper money, can create as much inflation as they want. It’s just a question of how determined they are.
In fact, there are some countries in the world, like Venezuela and Zimbabwe, that have had very high inflation — much too high, obviously. The Fed is targeting inflation at 2 percent and the reason they are not hitting that target is they’re setting their policy instrument at too contractionary a level to hit the 2 percent target.
The mistake in 2008 was not cutting interest rates quickly enough to try to stimulate the economy. The mistake in recent years was that the Fed tightened policy too soon and assumed that we were about to hit 2 percent inflation so they could start raising interest rates and ending Quantitative Easing. Well, it turned out we weren’t about to hit 2 percent inflation so they prematurely tightened policy; they tapped on the brakes before they needed to and that’s why we’re still a little bit under 2 percent inflation.
How Does Money Work, and What Makes it Special?
Charlie Deist: Let’s look under the hood at the transmission mechanism of monetary policy. For a lot of listeners it will be unclear how the Federal Reserve’s actions actually translate into inflation, but it’s easy enough to understand the quantity theory of money — the idea that money is neutral in the long run. Your blog, The Money Illusion, is named for the fact that there are all of these strange things that happen when we introduce money into the marketplace. In theory, it should be a neutral commodity: if we have twice as much money, then the prices should be twice as high. But in reality, there are frictions — prices don’t adjust so quickly, so we get all of these disequilibriums.
Why don’t we talk about how the Federal Reserve increases the money supply and what barriers there are to increasing the money supply in a way that would lead to some kind of steadily increasing nominal GDP target.
Scott Sumner: Let’s take that in stages.
First of all, what makes money special? Some economists believe money determines inflation. In fact, why are central banks given the job of targeting inflation? Why isn’t Congress doing it? Or someone else? And the reason for that is very basic. Money is special because prices are measured in terms of money. As an example, a hundred years ago, prices were measured in terms of gold, so gold was a very special commodity at that time. Supply and demand for gold would have an effect on prices and the economy. Now we don’t have a gold standard anymore — we have a paper money system. Because prices are measured in terms of money, the price level essentially becomes negatively related to the value of money. In other words, to put it in simple terms, if the cost of living doubles, the value of every dollar bill falls in half. So, who controls the value of dollar bills? Well, it’s the institution that has a complete monopoly on creating money, and that’s the Fed.
The Fed is the only institution allowed to create dollar bills, so obviously they’re going to be able to have an influence on the value of money by how much money they decide to create. If it were not true that printing a lot of money would ultimately lead to inflation, then it would be easy for poor countries to get rich. They could just print lots of money, give it to their citizens, and they’d all be millionaires. But everyone knows you can’t simply print money, give it to people, and have everyone be rich. You’d end up like Zimbabwe or Venezuela, with hyperinflation. So, then we clearly know that at some level printing lots of money creates inflation, and we know at some level the central bank — because they have a monopoly on printing money — is able to influence inflation.
The harder question is, “So, why did the Fed have so much trouble after 2008?” This is the puzzle that occurs when the Fed does something like Quantitative Easing (QE) — which is printing money basically — but doesn’t get much inflation out of it. What is special about this case is that interest rates fell to zero, and when interest rates are zero, money no longer has as much of an inflationary effect as in normal times.
Just think about it from your own personal life. The money created by the Fed doesn’t pay any interest at all (at least until recently). Even today, cash doesn’t earn any interest. If interest rates were 5 percent, would you want to hold a lot of an asset that doesn’t earn any interest? People typically don’t want to hold a lot of cash when interest rates are 5 or 10 percent. They could do better on other investments. But when interest rates fall to zero on, say, government bonds, then cash becomes just as good an investment as some of these other investments out there, and people suddenly are willing to hold lots of cash and, more importantly, banks are willing to hold a lot of what are called bank reserves.
This means that to a much greater extent than before the money the Fed pushes into the economy doesn’t get spent pushing up prices, it just gets held as an investment. That’s why the three QE programs did not create the full inflation that we see in Venezuela. Basically, interest rates were close to zero, and that money was being held as an investment because there was nothing better out there. Each time the Fed makes a large purchase of securities, it puts more money into the economy.
In theory, this should raise prices, unless that money is “sterilized” by zero interest rates or a policy of interest paid on cash held by banks as reserves. Unfortunately, it got a little more complicated after 2008 because the Fed began paying interest on bank reserves, so that creates a second reason why the money wasn’t creating inflation. They were actually paying banks to hold reserves, and that prevented the money from having its normal effect on spending. Essentially, printing lots of money is only inflationary if that money gets spent like a hot potato — passed from one person to another. But if the money is held under mattresses, in people’s homes, or in banks as bank reserves, it does not have the normal inflationary effect.
Bad Money Leads to Worse Government
Charlie Deist: So, it matters where the money is entering the economy and whether the first recipient of that money is choosing, in turn, to spend or lend it out. Around 2008–2009 the conventional wisdom held that monetary policy had lost its effect, but there are other ways of injecting money into the economy. In Venezuela or Zimbabwe, they usually have imbalanced fiscal policy — spending in excess of what they’re taking in a through taxation — so they print money to pay government employees or build “public works,” and this money gets into people’s hands, and then they go out and spend it.
Maybe I’m imputing bad motives where there are none, but it seems like there might be people in government who would prefer to engage in big stimulus projects than follow your solution of easing monetary policy. Fiscal policy provides a convenient excuse for people who would like to see their personal pet projects built. Do you think that could explain why Christina Romer — who was an economic advisor to Barack Obama — was ignored when she suggested other actions the Fed could take?
Scott Sumner: It’s probably a mixture of people being misguided and wishful thinking. There definitely is a feeling out there that monetary policy is not very effective at zero interest rates. I think that’s wrong.
The second problem is that there are a lot of people that actually favor fiscal stimulus. They viewed this as an opportunity since they favor government programs they’d like to see expanded. In other words, if monetary policy wasn’t doing the job, why not boost government spending?
One of the reasons I have fought hard for more effective monetary policy is that if we have an effective monetary policy, it actually makes other parts of our policy more efficient. We’re less likely to do wasteful things in terms of spending, bailouts, regulations and so on if the market economy seems to be working pretty well on its own.
It’s when you fall into a depression that there are all of these calls for the government to become heavily involved with either spending or bailouts. That’s where you get inefficient policies being substituted for monetary policy, which is more effective. In other words, if monetary policy keeps nominal GDP growing at 4–5 percent a year, there won’t be these calls for fiscal stimulus, or bailing out General Motors or big banks or things like that.
Charlie Deist: Especially from a libertarian point of view, there’s this temptation to look at anything that a central bank does as inherently evil and resulting in no good for the aims of a free economy. But you’re arguing, and Milton Friedman was trying to argue in a previous era, that the worst kinds of government policies tend to come up in periods when the monetary system gets mixed up.
Scott Sumner: I’d like to give a couple examples for your listeners that might be more conservative or libertarian: the U.S. in the 1930s and Argentina around the year 2000. Both countries made essentially the same mistake. In both countries, they had what you’d call a very contractionary or “hard money” policy; that is, a policy aimed at keeping inflation very low and not doing any sort of monetary stimulus. In both cases they overshot and went too contractionary, which pushed the economy into a depression.
In the U.S. it was the Great Depression, and in Argentina they also had a very deep depression. In both cases, the political system completely rejected the whole conservative free-market philosophy and swung very hard in a more interventionist, big government, socialist direction, because the depression was viewed as discrediting free market economics. The irony here is that the conservatives got the contractionary monetary policy they seemed to favor, but they ended up shooting themselves in the foot because the broader effect of that was to discredit free market economics.
What Milton Friedman wanted to do, and what I’m trying to do (which is quite similar) is advocate a sound monetary system in the hopes that it will make it easier to advocate free market policies. If the economy seems to be working pretty well, people are more comfortable with allowing free market policies as compared to when there’s massive unemployment and other problems. That’s when they turn to more socialist policies.
Never Reason From an Interest Rate Change
Charlie Deist: In the same vein, low interest rates are sometimes used as an excuse for fiscal stimulus. Recently there’s even been talk of finding a way to implement a policy of negative interest rates. It seems like that would be much more of a disaster for a free economy than what you’re arguing, which is modestly higher inflation in periods when nominal spending dips.
But how does changing the central bank’s “instrument” or the target actually gives them the traction for getting out of the “zero lower bound”?
Scott Sumner: This one is kind of tricky, because the short-term effect is often the opposite of the long-term effect with interest rates. Since it’s tricky, I’ll give you an example. Let’s go back to 2011. At that time Ben Bernanke was promoting an expansionary monetary policy in the U.S. — programs like QE. The European Central Bank (ECB), which is a little more conservative than Bernanke, was advocating more contractionary policy. They were worried about inflation and they raised interest rates several times in 2011, reflecting that more conservative stance. Now, go forward say five or six years. What happened? The U.S. gradually recovered, and then the Fed started raising interest rates. In Europe, exactly the opposite happened. Yes, in the short run, the ECB raised interest rates in 2011, but that pushed them into a deeper recession. So, Europe actually suffered much more than the U.S. during this Great Recession, and as a result of falling into a much deeper recession than the U.S., interest rates in Europe fell all the way to zero, and then actually slightly negative. They’re still basically right around zero or slightly below zero.
What looked like a more contractionary conservative policy in Europe of raising interest rates to prevent inflation ended up in the long run resulting in much lower interest rates than the U.S., which is now experiencing rising rates because our economy is much stronger than Europe’s.
Here’s my point: The only durable way to get higher interest rates and to prevent negative interest rates is to have steady growth in nominal spending in the economy — at least 4–5 percent nominal GDP growth. If you can do that consistently, that will allow you to have positive interest rates for savers and so on. You get into trouble with these zero or negative interest rates when you let nominal spending fall to very weak levels, or even negative levels — that’s what puts downward pressure on interest rates.
Interest rates are lowest in countries like Japan, which has had a recent history of zero or even negative inflation rates. I can’t emphasize enough that even though it looks like raising interest rates would be a good solution if you want higher interest rates, that can actually push you into recession that leads to lower rates in the long run.
If you want higher interest rates over time in a sustainable way you have to first get the right monetary policy. Then you have to basically tell the central bank, “Do what is necessary to hit that target.” Once you do that, it will gradually bring interest rates back to a more normal level from the extremely low levels we’ve seen in recent years.
Charlie Deist: My head gets tied into knots when thinking about how market expectations come into play. It seems like in order for the Federal Reserve to generate some amount of inflation or increased spending, they need to be able to credibly commit to keeping the money in the economy once it’s found its way in initially. As a lay person, I’m not usually paying close attention to indicators like interest rates or price indexes. Who are the first people to take note of what the Federal Reserve is doing, and how does that ripple outwards so that you eventually get increases in prices or more rapid spending through the economy?
Scott Sumner: The first people to notice these changes happening are in the financial markets. In economics, we have this theory called rational expectations, which holds that people have rational views on where the economy is likely to go given all the information they are provided. But this theory actually works best for the financial markets. In other words, most average people don’t pay much attention to what the Fed is doing, and so it doesn’t affect their expectations directly.
I think it tends to work more like this: if the central bank does something that’s actually effective and likely to be sustained over time, it shows up in the financial markets. If your listeners follow CNBC, they might have noticed that a Fed announcement can often make the stock market go up or down quite a bit (and not just the stock market, but the bond markets and the foreign exchange markets). They all move on news about Federal Reserve policy.
For instance, when the Fed announces a major QE program, that might make stocks go up, the dollar might fall on the foreign exchange market and so on. Then those changes spill over to the broader economy. The average person — who may not even know what the Federal Reserve is — may see the stock market change. That might affect their expectations, or when the exchange rate changes, that can affect our exports.
Of course, when interest rates change that affects the mortgage industry.
So, there are many different mechanisms. It’s the financial markets that can be thought of as the transition between what the Fed does and how it affects average people in the economy. So, the expectations of people in the financial markets who watch the Fed closely are very important in terms of how this process plays out.
Golden Fetters: The Limits of the Gold Standard
Charlie Deist: In the book you look at the reaction of asset prices and the stock market to different things going on — not just actions that the Federal Reserve is taking, but also things like the private demand for gold. If people are hoarding more gold, that will tend to have a deflationary effect. During the Depression it seems like that was the main problem. Everyone wanted to hold onto safer assets, but there’s this paradox of thrift, where when everyone simultaneously tries to save it results in lower incomes for everyone, so you need to have a fall in prices in order to equilibrate markets. But you write about how policy makers starting with Hoover (and continuing with Roosevelt) instead did the opposite. What did they do and how did that kind of backfire?
Scott Sumner: First, the value of money is the inverse of the price level, so under a gold standard anything that makes gold more valuable actually puts deflation into the price level. A gold standard actually makes it harder to do monetary policy because you don’t have complete freedom to just do whatever you want. You have to continue to make sure that the dollar is backed by gold at a fixed price and that there’s enough gold to redeem all the dollars out there and so on. That’s viewed as a constraint on the central bank.
Even given that constraint, the Fed quite frankly didn’t do a very good job during the Great Depression. The job they had to do was complicated by the fact that they were part of an international gold standard. To some extent, the Great Depression was an international phenomenon. That is, we had prices falling sharply all over the world. That partly reflected weaknesses in the gold standard that were created by World War I. I won’t get into all the complexities, but the bottom line is that it’s harder to do monetary policy under a gold standard.
Some people even advocate not having a Fed — just letting the gold standard run on its own in a free market system, which is one option. However, today we have this fiat money system under which the Fed doesn’t really have an option of doing nothing anymore. So even if you’re a free market purist who doesn’t like government intervention, given that we have a paper money system, the Fed has to adopt some sort of stance or strategy by default, whether it be controlling interest rates, controlling the money supply, or fixing the exchange rate.
Assuming that there is going to be a Fed and assuming that they are going to have some sort of policy — if only by default — for me that would be nominal GDP targeting , i.e., keeping total spending growing at around 4–5 percent a year.
Under the gold standard, back in the Great Depression days, the Fed had some ability to control things because the U.S. was such a big country even then. However, it didn’t have the unlimited ability to increase or decrease the money supply that it has today and therefore it faced a difficult problem during that period. What happened during the Great Depression is falling prices, which was very bad for companies and jobs and so on.
The question of “why did prices fall?” is really asking “why did gold become more valuable?” It became more valuable for a very simple reason. There was more demand for gold, and there was more demand for gold from almost everybody. The first mistake was the Fed increased its demand for gold. You recall there was a big boom in the stock market in 1929. They tightened monetary policy — they accumulated gold to try to kill the stock market boom. The Fed was worried that it was a bubble, so they tried to pop it by a tight money policy, accumulating gold. Unfortunately, they succeeded in popping the stock market bubble, but overshot into too tight of policy, pushing us into a depression.
Once we were in a depression it was like a perfect storm, with one bad event after another. Banks started failing, so people were hoarding cash. Banks needed more gold to back up that cash. Then people began to fear devaluation of the dollar — which did eventually occur by the way — and so they started hoarding gold, figuring gold would have more value if the dollar was devalued.
So essentially you had a series of mistakes starting with central banks accumulating too much gold. That pushed prices lower. Banks got into trouble and people started hoarding cash. That pushed prices even lower. With each of these steps, there was more and more demand for the ultimate form of liquidity — gold. For people interested in history, things really got thrown out of line by World War I, which pushed us into an unsustainable global system. But the gold standard does have this potential weakness. The hoarding of gold is deflationary and can create a depression.
Under our current system, that is less of a risk because if there’s hoarding of money, the Fed can adjust supply the money as needed to meet that demand.
Essentially the quantitative easing programs were intended to supply extra liquidity to prevent a repeat of the Great Depression. But while you can print money, you can’t print gold. When everyone wanted to hoard gold, there was no way to provide more.
Of course, the plus side of the gold standard is you don’t get high inflation because the you can’t print gold. Inflation rates tend to be very low under a gold standard, but it doesn’t give you the flexibility of dealing with a period of deflation. That’s why we got in trouble in the thirties.
Trust, Credibility, and the “Via Media” Between Hayek and Keynes
Charlie Deist: Milton Friedman made some of these arguments, and a lot of them were incorporated into the conventional wisdom, which holds that we can improve outcomes if we have some sort of stable rule. But now you have conservative-leaning economists including the Austrian school on one side, and then the mainstream economics profession on the other side, both insisting that there is not any continuing role for monetary policy in the current environment.
Let’s start with this question of the Austrian school. During the Great Depression, the dollar devaluation people were worried about did eventually happen. You tend to praise the dollar devaluation for at least temporarily getting us out of the Depression, but couldn’t that be used as an argument that the government is whimsical, and its decisions around monetary policy can’t be trusted?
In general, the public is afraid of a massive devaluation because it hurts savers and things like that. So, what would you say to an Austrian?
Scott Sumner: Well, they should be worried about too much of that. I’m going to have to oversimplify a little — apologies on that score, especially to the Austrians. But it might help to contrast Austrian and Keynesian ideas. I think of them as two extremes. At the risk of oversimplification, the Keynesians worry a lot about times when there’s too little spending in the economy — when there are recessions and high unemployment — so they are known for favoring stimulus to prevent those negative outcomes. Austrians are known for worrying about the opposite problem of too much spending — excesses in the economy, bubbles, unsustainable investment, booms created by easy money — those sorts of things.
Now, smarter people on both sides worry about both, but that’s the way I think of the difference between Keynesian and Austrian. It’s a difference of emphasis, on whether you’re worried about excesses on the “too much spending” side or excesses on “too little.”
I think of Keynesians as wanting to create flat land by filling in the Grand Canyon, and Austrians as wanting to create a flat land by leveling mountains to make it flat.
My view is that both problems are equally worth worrying about. I try to focus on a stable path of spending. That is, you don’t want too much spending like we had in the 60s and 70s — creating very high inflation — or too little spending, like in 2008 and 2009, or in the Great Depression, creating high unemployment. You just want spending to grow at about 4–5 percent a year. I see the root cause of the problem as not being just a question of too little spending at some time or just a question too much, but rather the instability — the cycles — the movements of high spending in one year and low spending and another. You want a policy that stabilizes the monetary system and the amount of spending in the economy.
A lot of people worry about credibility and trust and so on. I would emphasize two points there. Yes, it’s very important to have a credible monetary system where there’s trust that governments will keep their promises. However, you have to make sure that the promises you make can be kept. So, in an emergency, are you actually willing to stick with your promise?
I mentioned Argentina in 2000, and also America in the 1930s — in both countries, conservatives had made promises that should not have ever been made, because they couldn’t be kept. In the 1930s it was, “We’ll keep the price of gold fixed.” Argentina’s mistake was, “We’ll keep the Argentine peso fixed one-for-one to the U.S. dollar.” It was called a currency board.
Those were very strong promises that the governments found politically impossible to keep once the countries got into a depression. Keeping that promise meant very high unemployment year after year, which the voters just wouldn’t accept. To some extent, Greece’s decision to join the Euro is a similar type of mistake, except Greece is continuing to stay in the Euro because it’s almost impossible to leave once you join. But that’s a different story.
Here’s what I suggest in terms of credibility, trust, and promises:
First, it’s important to set up a strict rule that the central bank has to follow.
Second, it’s important to set up a rule that is actually feasible — one that you can keep following even in an emergency, recession, or depression, because it won’t hurt the economy. You have to make sure you get your policy commitment right. I don’t believe that pegging the price of gold under a gold standard is a commitment that that a modern government will be able to keep. Therefore, I worry about a crisis where there is a loss of trust, and people start to fear the promise won’t be kept. So instead of trying to go back and recreate what we had in the 19th century, it’s better to set a system that is based on a promise that the government actually should be able to keep.
For people that are cynical about this, let me point out that despite all the problems that the U.S. has faced in recent years, and despite the criticisms of the Fed — which are well-deserved to some extent — there is one policy that the Fed has done a reasonably good job of adhering to, and that was their decision a few decades ago to keep inflation close to 2 percent. Ever since about 1990, the Fed has kept the inflation rate pretty close to 2 percent — it has fluctuated between 1 and 3 percent. That was by design — it didn’t just happen naturally. Prior to 1990, there was never any sort of a tendency for inflation to always be 2 percent — it would usually be much higher or much lower. The Fed has made mistakes in recent years, but one thing they continue to do reasonably well is keep the inflation rate close to 2 percent.
That means that people who want to make long-term plans, like borrowing money to buy a house, have some idea of what the price level will be like decades out in the future when they repay that loan. That was not true as recently as 1960 or 1970. Nobody borrowing money in 1960 or 1970 had even a clue as to what the cost of living would look like 20 or 30 years later, because we had years of 5, 10, 13 percent inflation.
Since 1990, inflation has averaged 2 percent. Even without a gold standard, people shouldn’t be too cynical. Price stability is still possible, and inflation targeting keeps the inflation rate pretty low and stable. Again, I happen to favor nominal GDP targeting, but if we’re going to do inflation targeting, that’s something the Fed can achieve if it’s determined to do so — not perfectly, every single year (it’s been a little below two percent in recent years), but roughly 2 percent.
The Future of Monetary Policy: Futures Targeting and Cryptocurrency
Charlie Deist: The uphill battle of shifting from inflation targeting to a nominal GDP targeting regime might involve appeasing inflation “hawks” who say that any departure from this 2 percent rule could result in runaway inflation. But it seems like in recent years the problem has been that the Fed has been far too timid in not signaling a desire to get back on the path of 2 percent inflation per year. Something to emphasize from a public relations standpoint is that nominal GDP is synonymous with nominal income — income equals expenditures. An increase in prices also translates into an increase in wages and incomes.
In just the last couple of minutes, I want to go back to your process for writing the book, The Midas Paradox, which involved going through the New York Times archives and seeing how statements from the Federal Reserve impacted asset prices. Today we’re seeing a lot of fluctuations in the stock market. I’m curious to get your take on how these relate to monetary policy, and if there was anything else of interest that you found looking through New York Times archives in this period of the Great Depression through the beginning of World War II. Was there anything particularly interesting — economics-related or otherwise — that caught your eye?
Scott Sumner: What caught my eye most is how efficient financial markets are at processing information. I was continually amazed at how the financial markets were ahead of the rest of the world in understanding what was going on — ahead of the economists, the politicians, and the journalists. So often there would be some shock to the economy — a policy change, or a crisis — and you’d be able to observe the financial markets respond very quickly to this news. I found that the responses seemed very rational based on what we know now, but at the time people were often bewildered. They were wondering, “Why is this happening?” They didn’t see the mistakes that were being made by the Federal Reserve and others, but the financial markets did see these mistakes. Indeed, one reason for the 1929 stock market crash was that the market saw the Fed making a serious mistake with its tight money policy.
That newfound appreciation of the financial markets in terms of their ability to see what is going on with monetary policy has led me to become what some people call a “market monetarist.” That is, instead of having a committee in Washington setting interest rate targets, I favor a system where monetary policy is set automatically at a level where the market thinks we’re likely to have 4–5 percent nominal GDP growth. If you look over time, the markets are the most efficient mechanism we have for processing information.
Also, a market-based approach would reduce uncertainty. Right now, the markets are often trying to figure out what the Fed will do. I don’t know if that was behind the recent stock price drop off a few weeks ago, but I do know that quite a few of the major movements in stock prices over the years have been in response to uncertainty about what the Fed was doing, or even disapproval of a decision made by the Fed at a particular meeting.
If the market were directing monetary policy, there would be less uncertainty and the economy would be more stable. People would be able to plan with more of a sense of confidence as to where things were going in the future.
Charlie Deist: You talk about market monetarism and tying the Fed’s actions to some sort of actual market indicator. People have even talked about a some sort of futures market for nominal GDP that is linked to the buying and selling of bonds, i.e., if the expected price of a certain asset that reflects inflation expectations — like the TIPS (Treasury Inflation-Protected Securities) — rises or falls above its target, that would automatically trigger some sort of action by the Fed. These futuristic ideas are interesting to think about and we’ll see if they gain prominence in the years to come.
Given that cryptocurrency has been a hot topic, do you have any insights into how that might factor into what the Federal Reserve is thinking of doing, or whether cryptocurrency serves as another hedge on dollar devaluation — something thing that will prevent the Federal Reserve from generating too much inflation in the future?
Scott Sumner: I said earlier that what makes money special is that things are priced in terms of money. Monetary policy is so powerful. When you change the value of the U.S. dollar, it changes the whole price level in the economy, and everything is impacted by monetary policy. Right now, cryptocurrencies are not used very widely in pricing. You don’t see price tags in stores measured in Bitcoin, and wages are not paid in Bitcoin, so at the moment the cryptocurrency market is not really interrupting the Fed’s ability to do monetary policy.
But I have to admit that I didn’t even see this coming. If you asked me 20 years ago about the idea of Bitcoin, I would have thought it was a little bit far-fetched. It’s obviously been more successful than a lot of economists like myself anticipated. It would really provide a challenge for monetary policy if it becomes so successful in the future that people start substituting Bitcoin for U.S. money and doing more and more business transactions in Bitcoin, and most importantly, if wages and prices start to be measured in Bitcoin. It’s possible the Fed would lose its influence over the economy. We’re not at that point yet and if we start to get there we’ll have to probably do some more serious thinking about what that means.
Right now the value of Bitcoin is very unstable. Ideally, you want your monetary system to have fairly stable value over time. So, if we go to this kind of currency replacing the U.S. dollar, we’ll have to come up with one that is able to maintain a fairly stable value over time so that it doesn’t cause a lot of instability in the economy from cycles and ups and downs and its value.
Charlie Deist: I think that is just about all the time that we have today. I’ve been speaking with Professor Scott Sumner, who is the Director of the Program on Monetary Policy at the Mercatus Center. You can check out his work at The Money Illusion, which is a blog that became popular during the period after the Great Recession and heavily influenced the monetary policy debate. You can also buy his book, The Midas Paradox: A New look at the Great Depression and Economic Instability, published by The Independent Institute. And we thank Scott for taking the time to talk with us and look forward to continuing to follow your work in the years to come.
Scott Sumner: Thanks so much, Charlie.
Z is for the Zeitgeist
Monetary policy is still much more of an art than a science. It depends too heavily on mysterious and unknown forces that are built into the complexity of the economy and the feedback loops between supposedly autonomous actors like the Fed and actors in financial markets. The Zeitgeist could be shifting from one in which central bankers carefully target interest rates to one where it follows a simple rule — like Milton Friedman suggested — and uses market forecasts to verify that it is on track to hit a flexible inflation or nominal GDP target. Or, perhaps the days of central banking are coming to an end.Only time will tell.
