Original Sin of the Macroeconomics

I’ve been reading articles on the Keynesian/NeoFisherian debate for a while. It’s an interesting debate for me. Before commenting on the debate, I’d better start from the Great Recession.

After the Great Recession, three “heterodox” objections arose in monetary economics – schools of thought that reject the mainstream models. In my opinion, all three objections are responses to a significant failure of mainstream academic models.

Consider the following claim in the mainstream model:

“The policy of lowering interest rates to low levels is very expansionary. Combined with a large fiscal stimulus package, it can lead to high inflation or a debt crisis.”

A policy based on this assumption has been implemented in Japan over the past thirty years. Still, there was no significant inflation and debt crisis. I think this is an anomaly that needs to be explained.
Let’s look at how these three new schools of thought deal with this situation:

  1. MMTs argue that there are no restrictions on how much the governments of countries with their own fiat money can borrow. They recommend that central banks in these countries reset their interest rates. Inflation will be a problem only if spending exceeds the productive capacity of the economy, in which case the proposed solution is high taxes.
  2. NeoFisherians claim that low-interest rates are not expansionary policies but rather a contraction policy that would lower inflation rates.
  3. According to Market Monetarists, interest rates are not a reliable indicator of the monetary policy stance. However, a long period of falling interest rates often indicates a tightening of monetary policy. Moreover, market monetarists argue that those who argue that interest rates are an element of monetary policy make too much reasoning from a price change.

So we have three new “heterodox” models, none of which fit the mainstream model, and none is the same. How will it end in such a mess? With other anomalies observed in Switzerland and elsewhere after 2008, the development of these models became almost inevitable. But why did these three new schools appear? The answer to our question lies in the fact that there were important divisions in the monetary economy even before 2008.

Those who preferred a more leftist approach to the Keynesian model tended to reject approaches closer to quantity theory on monetary policy strongly. In this sense, MMT can be seen as the most anti-quantity theory model ever developed. The foundations of MMT are the polar opposite of monetarism on a wide variety of issues.

A more right-leaning group tended to use a classical macroeconomic model (flexible prices). In these models, the Fisher effect was far more important than the liquidity effect. Therefore, it was natural that more classical-leaning economists would turn to the explanation of the Japanese anomaly emphasizing the Fisher effect: they argued that changes in inflation expectations were strongly associated with changes in nominal interest rates (NeoFisherians).

A group of economists from monetarist traditions was called market monetarists. They emphasized the importance of the wide range of links between money and interest rates, including liquidity, income, and Fisher effects. This group accepted that changes in interest rates are a kind of epiphenomenon of monetary policy change. They rejected the assumption that changes in interest rates could tell us everything useful about changes in monetary policy stance.

I blame both sides for too much reasoning from price changes. Both sides make too much inference from the change in interest rates as if interest rates were monetary policy. In fact, interest rates are not monetary policy: they are simply the prices of loans. And, because of the “never reason from price change” rule, it also makes no sense to talk about the effects of a change at any price.

What does this mean? In fact, the same rule applies to the amount of change. An increase in quantity can be associated with lower or higher prices, depending on whether it results from a change in supply or demand. If I told you that the base money would triple in 2021, what would your inflation expectation be? Would it raise or decrease? The correct answer actually depends on whether the base currency tripling is in the US or Zimbabwe.
So the question “Will QE cause inflation to rise?” is as absurd as the question “Is it likely that lower interest rates will lead to higher inflation?” Absolutely pointless question without context.

Most economists know this rule, so why are they constantly misjudging monetary policy? Frankly, this is a question that has been puzzling me for a long time.

When we see cigarette prices rising, we do not actually know whether this reflects more or less cigarette consumption because we do not know whether the price increase reflects increased demand or reduced supply. However, let’s say we learned why cigarette prices actually increase: The increase is due to a higher tobacco tax. In this case, we can predict that the price increase will lead to less consumption.

I think something like this is happening with interest rates for many economists. On a theoretical level, almost every economist can admit that high nominal interest rates can be the liquidity effect from tight money or the inflationary effect of loose money. But when the central bank changes interest rate targets, they believe they know one hundred percent of why interest rates are changing; then they believe there may be some kind of shortcut in analyzing the impact of that change in interest rates. But this doesn’t seem right because money is a special commodity.

Keynesians correctly assume that an increase in interest rates relative to the equilibrium interest rate tends to lower inflation. However, they mistakenly think that it causes the market rate to increase relative to the equilibrium rate when the target rate rises. This may be true, but it can also be false. Keynesians, I think, underestimate how central banks affect policy decisions and the equilibrium interest rate.

NeoFisherians correctly assume that if the Fed’s interest rate target and the equilibrium rate rise simultaneously, inflation will increase. Also, the assumption that, in the long run, the target rate and the equilibrium rate will tend to move together is justified. So far, so good. However, they calculate how often an increase in a given Fed interest rate will increase the interest rate differential (target rate minus the equilibrium rate) and therefore tighten monetary policy.

What’s important here is the target ratio minus the equilibrium ratio. While Keynesians tend to exaggerate the impact of any move in the target rate on this gap, NeoFisherians tend to ignore this distinction and think that the target interest rate moves with the equilibrium interest rate. NeoFisherians are right that the Fed could change the balance rate. But the Fed does this by moving the target rate in the desired direction. But the Fed does not do this by moving the target rate in the desired direction it does the opposite.

To get a better grasp of the situation, let’s consider two historical examples.
There were two main reasons for the Great Depression: an overly tight monetary policy caused the NGDP to fall between 1929 and 1933, and then a series of New Deal policies such as NIRA slowed the recovery. Most experts and policymakers looked at the rapidly falling level of nominal interest rates and assumed the money was easy. In fact, rates were falling due to the decline in loan demand caused by the depression itself. The money was actually very tight. This is a perfect example of the “never reason from price change” rule.

Btw, reasoning from a price change also caused the Great Recession. I’ll explain cover the example of the Great Recession in a long time to approach the model theoretically.

First, I will start with a simple model (without IOR) and add IOR to the model in the second step.
Let’s start with the following equation:

BmV = PY

Now we can look at the model:

MV (i) = PY

Since V is positively associated with i, low-interest rates are contractionary, reducing V and hence NGDP.

“Well, other economists didn’t know that, but you and the other market monetarists were very clever, so you discovered it?”
Indeed, they didn’t know, but there are some reasons for this:

  1. Sometimes (not always), the decrease in interest rates is caused by the increase in the monetary base (of course, this did not happen in 2007-2008).
    When an expansionary monetary policy is implemented, specifically when there is an external increase in (Bm), interest rates fall and in this case, (V) tends to have a lower rate of falling than (M).
    Therefore, even if the specific effect of low-interest rates causes NGDP to fall, the whole policy causes the NGDP to rise as a whole.
  2. Another problem stems from the Keynesian model.
    It would not be absurd to say that any Keynesian model in circulation that says low-interest rates are expansionary is wrong. This is probably why economists were so confused about the Great Recession.

Many people confuse aggregate demand with consumption. So they think that low rates encourage people to “spend,” which somehow increases aggregate demand and NGDP. But that’s not the case, at least not what they assume. If by “spending,” we mean higher velocity, yes, more spending increases NGDP, but we have seen before that lower interest rates do not increase velocity but rather decrease it.

Even worse, some assume that “spending” is the same as consumption: If lower rates encourage people to save less, aggregate demand and NGDP will increase. This is a tremendous misconception: if you don’t spend the savings, it goes to the investment, and the investment is part of NGDP. This is the Achilles’ Heel of the amateur Keynesians: at this point, they remember reading something about the paradox of saving and the attempt to save more can put NGDP under pressure, and eventually, people cannot save more, instead, NGDP will fall.
This is not quite right, but even if it is true, it does not affect my claim that low rates are contractionary.

To see the problem with this analysis, we must look at Keynesian explanations for increases in aggregate demand.
There are still things that macroeconomics cannot solve, and an important knot of this confusion lies in the concept of “animal spirit.”

Most economists do not understand what caused the recession in 2008, even a decade later (including the Great Depression for some). Most business cycles in the US have resulted from demand-side shocks, namely changes in NGDP. These demand shocks stem from the unbalanced monetary policy, i.e., expected changes in the future NGDP.

While most economists agree with my first claim, they disagree with the claim that the root of the problem is unstable monetary policy. Since these economists cannot identify unstable monetary policy, they follow Keynes and add a sloppy animal spirit to their model. Their models cannot explain why spending is falling, so they assume a kind of mysterious pessimism is developing in the minds of millions of entrepreneurs. But there is a problem here. Due to the law of large numbers, the total business and consumer psychology suddenly change for no reason. The problem is modeling policymakers: Can anyone claim that we really know exactly how the 12 members of the FOMC will behave?

So what is the Keynesian explanation about increases in aggregate demand? One is that animal spirits are driving businesses to invest more. Another is that consumer optimism encourages consumers to spend more. The other is that the fiscal policy expands and increases the budget deficit.

What do these three explanations have in common? In all three cases, the “shock” leads to higher interest rates (you can use the S/I chart).
Yes, high interest rates increase the velocity in all three cases, and hence high (V) leads to higher NGDP when the monetary base is fixed (i.e., ceteris paribus).
But here’s the problem: the above is an example of low-interest rates that increase aggregate demand, but not aggregate demand, and at the same time increase interest rates.

Let’s move on to the difficult part.
I argued that slightly higher interest rates were expansionary and low rates contractionary. But the higher IOR is really contractionary. Okay, what’s the difference between the two?

Let’s go back to our simple model above. We could apply this model in a non-IOR world. Now is the time to add the IOR:

BmV (i – IOR) = PY

So the velocity is positively related to the difference between the market interest rate and the interest rate on money. The difference is the opportunity cost of holding reserves.
To make monetary policy tighter, the Fed only needs to shrink the difference between i and IOR. This reduces the opportunity cost of holding reserves, which causes more demand for the base currency, which has a contractionary effect.

The strange situation we’re in is this:

  1. Almost everyone assumed higher odds were constrictive, but almost everyone was wrong.
  2. Then came the IOR, and the IOR is really contracting.
    To summarize, the IOR is not really a market price: the positive IOR is a subsidy on the base money, and the negative IOR is a “tax” on the base money.

Let’s do a thought experiment:
Let’s say the Fed announced that it was transitioning to a tight monetary policy just this second. Recession expectations cause a decrease in loan demand. The loan supply is also decreasing, but not sharply. Thus, nominal interest rates are falling. Low nominal interest rates reduce the velocity, leading to a decrease in NGDP(for simplicity, let’s assume no change in monetary base). As nominal wages are sticky, low NGDP causes recession. Fear of recession realizes itself and leads to recession.
But even if this fear did not fulfill itself, the Fed was willing to cut enough base money to cause a recession. But that was not even necessary; low-interest rates (by lowering V) took care of the situation on its own.

Let’s do the same thought experiment more concretely:
The Fed has switched to a fixed exchange rate system where the dollar will appreciate by 1% annually. This reduces the nominal interest rates in the US by 1% due to the interest parity condition.
Both policies create recession by reducing NGDP and both lower nominal interest rates. Real interest rates are not that important, as the distinction between inflation and real GDP is completely arbitrary (they have nothing to do with anything more tangible than “utility”).

I claim that such a thing happened in 2007-2008. The Fed has made it clear that it attaches more importance to the markets than NGDP. They signaled that they will allow NGDP growth to slow if the need to restrain inflation ceases. Markets saw what the Fed was doing, and they said: “We have seen that you care about inflation, we care about NGDP growth, not inflation. As your policy will reduce NGDP growth, we will reduce the Wicksellian equilibrium rate in anticipation of recession.”
This is exactly what happened in December 2007.

Later in 2008, the Fed signaled the markets that there was some kind of inertia in adjusting the interest rate to conditions. Markets saw this and concluded that the Fed could not cut rates quickly enough to avoid recession. This made the markets more pessimistic, which decreased the Wicksellian equilibrium rate even faster. This is exactly what happened in the second half of 2008.

Where do all these examples lead?

In 1936 Keynes invented a macro model that reflected the spirit of his time. This is the wrong way to do a macroeconomic model. If you are making a model, the model should be based on empirical data for almost all countries and all time.

If your model cannot explain why the United States experienced a major deflation (about 30% decrease in NGDP) from 1920 to mid-1921, your model is not working. If it can’t explain why industrial production suddenly dropped by 30% after the mid-1920, it wouldn’t work. Your model explains why the recession of 1921 followed a very rapid recovery, but if it cannot explain why there was an 8-year recovery in the Great Depression, your model is not working.

If your model is going to work, you need a “General Theory” that applies anytime and anywhere.
Unfortunately, most macro models cannot explain what I mentioned above and others. Do we have another explanation for the difference between the US and Hong Kong Phillips curves? Of course, there is no such simple and elegant alternative explanation.

As in the example above, the market monetarist model is a fairly simple yet incredibly powerful framework for analyzing various macro issues.
Of course, we still need to model all the factors that affect aggregate supply. However, this framework can easily explain the problems I mentioned at the beginning.

The large decline in NGDP during 1920-21 was mostly due to a decline in the monetary base. The rapid recovery was due to nominal wages being more flexible in those days. It should be added that the monetary base is not visible even on most NK models. What are the explanations for 1920-21?

This framework is also valid for modern times. The reason for the recessions in 1982 and 2008 was that the current NGDP growth fell far below expectations due to tight monetary policies. As I mentioned in the first post of this blog, recessions are always and everywhere monetary phenomena.

Going back to the beginning, the error of equating changes in interest rates with monetary policy seems to be the original sin of the macroeconomy. It caused a lot of confusion, recession, and all of this led to the birth of the three schools of thought I mentioned.

Later, I will write a series of posts where I analyze NeoFisherianism, MMT, and NeoKeynesianism one by one.

Notify of
Newest Most Voted
Inline Feedbacks
View all comments

[…] Original Sin of the Macroeconomy […]


Start with IOR=0% and Fed Funds=9%.

Next, move to IOR 2% and Funds 7%.

What happens?

[…] Commenter Arda asked the following question: […]


So, do you mean cash here? I think cash demand is extremely inflexible for everyone in a modern economy. Lowering interest rates doesn’t cause people to rush to bank accounts to withdraw money.

[…] caused recession, it is misunderstood. So let me try to clarify this phrase in explaining for the second time why both Keynesians and NeoFisherians are […]