Category Archives: Scott Sumner is Coherent

Finance and Macro

Posted on by

Nick Rowe writes,

Here’s a very simple (and totally inadequate) theory of the rate of interest: it is set by the Bank of Canada. Add or subtract adjustments for risk, duration, liquidity, and earnings growth, and you get the equilibrium earnings yield on stocks. Take the reciprocal, and you get the P/E ratio. Done.

Why is that theory totally inadequate? Because the Bank of Canada does not set interest rates in a vacuum. It sets the rate of interest it thinks it needs to set to keep inflation at the 2% target. And that interest rate in turn depends on things like the demand for goods, and the Phillips Curve, and on the inflation target. And the demand for goods in turn depends on things like desired saving and investment, both in Canada and around the world. And those in turn depend on time-preference, and expectations of future income, and on the marginal rates of transformation of present goods into future goods, and whether there will be a demand for those future goods or a recession.

Read the whole thing.

This first paragraph reminds me that I have meant to write an imaginary Q&A with Scott Sumner.

Q: Why did the stock market go up about 2 percent the other day?

SS: Because the Fed announced an expansionary policy.

Q: But the Fed announced that it was tapering its purchases of assets, although they issued a “forward guidance” that interest rates would remain low. Given the somewhat contradictory announcement, how do we know that it was expansionary?

SS: Because the stock market went up about 2 percent.

Nick’s second paragraph reminds us that central bank policy is also endogenous. That is the way that I think of it.

So what’s my explanation for the rise in the market, which was obviously in response to the Fed announcement? A couple of possibilities.

1. Perhaps they read the taper announcement as an indication that the Fed has information that the economy is doing well. They took it as good news.

2. Perhaps a few key Wall Street gurus interpreted the announcement as good news, because of (1) or because they are devoted followers of Scott Sumner or because of the meds they were on or whatever. Then everyone else realized that if the gurus were optimistic then stocks would go up, so they pushed stocks up. It was collective irrationality. As Fischer Black famously said, the stock market is efficient only to a factor of 2.

The way I reconcile finance with macro is that I minimize the weight I give to macro. Markets are happy to let the Fed wiggle around an interest rate or two, as long as it does not wiggle too hard on an interest rate that really matters to the economy. If the Fed were to wiggle too hard on a rate that matters, the markets would find a way around that particular part of the money market in order to make that interest rate matter less. As an economist, your best bet is to treat interest rates and stock prices as determined by financial markets, rationally or otherwise (I vote otherwise), and not by the Fed.

Normal AD vs. the Credit Channel

Posted on by

‘Uneasy Money’ writes,

try as they might, the finance guys have not provided a better explanation for that clustering of disappointed expectations than a sharp decline in aggregate demand. That’s what happened in the Great Depression, as Ralph Hawtrey and Gustav Cassel and Irving Fisher and Maynard Keynes understood, and that’s what happened in the Little Depression, as Market Monetarists, especially Scott Sumner, understand. Everything else is just commentary.

Pointer from Tyler Cowen. This argument broke out five years ago, and it is no closer to being settled. I might phrase it as the following multiple choice question:

a) the economic slump caused the financial crisis (the Sumnerian view, endorsed above)

b) the financial crisis caused the slump (the Reinhart-Rogoff view; also the mainstream consensus view).

c) both are symptoms of longer-term structural adjustment issues (I am willing to stand up for this view. Tyler Cowen also is sympathetic to it. Note that I do not wish in any way to be associated with Larry Summers’ view, which is that the structural issue is that we have too much saving relative to productive investments.)

d) both are symptoms of a dramatic loss of confidence. As people lose confidence in some forms of financial intermediation, intermediaries that are heavily weighted in those areas come to grief. Se see disruptions to patterns of trade that depend on those forms of intermediation. Moreover, as businesses lose confidence, particularly in their ability to access credit, they trim employment and hoard cash.

I want to emphasize that I see a reasonable case to be made for any of these views. There may be yet other points of view that I would find reasonable (although Summers’ “secular stagnation” is not one of them). In macroeconomics, if you think you have all the answers, then I cannot help you. I think that this is a field in which doubts are more defensible than certainties.

Normal Macro and Financial Crises: Road Friction and Flat Tires

Posted on by

Economists these days seem to want to describe two financial regimes. One is a “normal” regime, which can be regulated by conventional monetary policy. The other is a “crisis” regime, which cannot. Here is a metaphor to think about:

Imagine the economy as a car, with Fed monetary policy consisting of pressing on the accelerator pedal. The Fed tries to maintain a constant speed, and sometimes that means pushing hard on the accelerator and other times it means letting up. In the normal regime, the Fed just deals with road friction and hills.

In the crisis regime, the car has a flat tire. The Fed can press really hard on the accelerator pedal, and yet the car is not going to go as fast as people want.

Until recently, macroeconomic theory focused on the normal regime, dealing with road friction and hills. What are the microfoundations? What policy rules work best? etc.

But what about the crisis regime? Scott Sumner’s view would be that there is no such regime. He would say that we are seeing a normal regime in which the Fed has stubbornly failed to press hard enough on the accelerator pedal.

When I hear mainstream economists praise TARP, I think they believe that troubled banks were the macroeconomic equivalent of a flat tire, and you needed TARP to patch the tire. Perhaps this is correct. However, I do not think we can tell this story very well by using the models that were designed to describe the normal regime. As of now, I think that there is a huge gap between mainstream intuition, which thinks in terms of the flat tire, and mainstream modeling, which deals with road friction. In particular, treating financial markets as if they were just another potential source of road friction is probably not going to cut it.

I also do not think that “the zero bound” is the flat tire.

What might be the flat tire is an adverse equilibrium in an economy with multiple equilibria. Think of the economy as having channels of trust. When the channels of trust are open, we have the good equilibrium. When the channels of trust are closed, we have the bad equilibrium. Finance is particularly subject to these multiple equilibria. If people believe that financial instruments are safe, then borrowers can obtain lenient credit at low rates. But in an adverse equilibrium, creditors have doubts, and borrowers are constrained.

In the adverse equilibrium, there is less economic activity. This might explain the intuition that the financial crisis was horrible, and bailing out banks kept things from getting worse. However, it does not necessarily support the intuition that fiscal policy and quantitative easing are helpful.

And I do not necessarily endorse this particular model of the flat tire.

Secstag: All Things to All People?

Posted on by

Daniel Davies writes,

The US economic policy structure was aware that they were accommodating China and NAFTA, and aware that the tool of demand management was consumer spending. They might or might not have been aware that the consumer spending was financed by borrowing against housing wealth, but if they weren’t, they thundering well should have been. They got a structural increase in personal sector debt because they wanted one and set policy in order to create one. There’s no good calling it a “bubble” or a “puzzle”

Pointer from Tyler Cowen. Read Davies’ entire post.

We have the basic identity

S – I = (T-G) + (X-M)

That is, the excess of domestic savings over investment equals the government surplus plus the trade surplus. This is true whether we are in a recession, a boom, or anywhere in between.

What Davies seems to be saying is that China wanted a lot of (X-M), which gave us a big negative (X-M). Holding (T-G) constant, this gives us a big negative S-I. Since we didn’t do much I, we did a lot of dissaving. And this drop in personal saving is yet another meaning for the very plastic phrase “secular stagnation.”

Oy. Scott Sumner comments,

There can’t be a structural shortage of demand, because demand is a nominal concept.

For decades after The General Theory, there were arguments over what Keynes really meant. Seeing what Larry Summers has unleashed, one can understand how this happens. At a time when economic performance is disappointing and people are groping for explanations, a guy who is known to be a great economist offers an answer that is vague but sounds clever. He then leaves it to other people to come up with a precise version. Unfortunately, the precise versions are problematic, meaning that they are either unsound in terms of theory, inconsistent with evidence, unable to support the explanation and policy implications of the vague version, or all three. We proceed to cycle back-and-forth between the clever-sounding vague version and the precise, problematic versions.

Why Interest on Reserves?

Posted on by

Scott Sumner writes,

Back in late 2008 a few money market funds got into trouble and were in danger of “breaking the buck.” That’s due to their policy of pricing each share at $1. The solution is to allow the price to fluctuate. The Fed should have given the industry 6 months to prepare for negative interest rates. Instead they bailed them out and propped up interest rates at 25 basis points, in order to insure they would never break the buck.

If not for the money market industry the Fed could have already cut the fed funds target to around negative 0.25%, and the same for the interest rate on reserves. In that case (and assuming the IOR also applied to vault cash) it’s likely that most of the ERs would exit the banking system and end up in safety deposit boxes. But three trillion dollars is a lot of Benjamins, and despite the cash hoards you observe in places like Japan, a more likely outcome would have been hyperinflation. Obviously that would not be allowed, so what this thought experiment really shows is that with that sort of negative IOR the Fed could have gotten the stimulus it wanted with much less QE.

The decision to pay interest on reserves is one of the great mysteries of the 2008 response to the financial crisis. In terms of monetary policy, it is clearly contractionary, and a financial crisis seems like an odd time to engage in a contractionary policy.

The Fed acts in mysterious ways. At the time, the Fed said,

Paying interest on excess balances will permit the Federal Reserve to provide sufficient liquidity to support financial stability while implementing the monetary policy that is appropriate in light of the System’s macroeconomic objectives of maximum employment and price stability.

Which to me says exactly nothing. It could be that the only way to find out the basis of the Fed decision is with an audit.

Scott re-interprets Larry

Posted on by

Scott Sumner writes,

Summers claims that some sort of exogenous shock has reduced the long run real interest rate on safe assets.

That is not what I heard. Go back and listen to Larry’s response to Jeff’s question. Larry is talking about a savings glut and a decline in the cost of physical capital.

Think of an economy with three assets: money, risk-free Treasuries, and physical capital. What I heard Larry saying was that because of the savings glut and the low price of computers, the full-employment real return on physical capital is negative. That is a silly idea and a false idea, but that is what I heard Larry say. Ben Bernanke heard the same thing.

What Scott heard Larry say was that the demand for risk-free Treasuries is high, but the demand for physical capital is not so high, so that there is still a sizable risk premium on risky assets. My comments:

1. That may be a good story for, say, the fourth quarter of 2008. Larry claims to be talking about a decades-long phenomenon, pre-dating the crisis and continuing into the indefinite future.

2. The policy implications of that story are somewhat different than the policy implications of “secular stagnation.” If there is too much savings, then Larry wants to argue that the government needs to use up the savings. If what is holding back investment is high risk premiums, then more government spending is not such an obvious remedy.

Gerald O’Driscoll vs. Scott Sumner

Posted on by

O’Driscoll writes,

When the Kennedy and Johnson Administrations started engaging in fiscal activism under the sway of Keynesianism, the Federal Reserve under Chairman William McChesney Martin monetized the resulting deficits.

Sumner writes,

I do not believe the “Great Inflation” of 1965 – 81 was caused by the monetization of fiscal deficits. The deficits were relatively modest during that period, and the national debt was falling as a share of GDP. Deficits became a much bigger problem beginning in 1982, but that’s exactly when inflation fell to much lower levels. Instead the Great Inflation was probably caused by a mixture of honest policy errors and politics.

Let me throw a third hypothesis into the mix. There was a fair amount of money illusion in financial markets in the 1970s. That is, people looked at high nominal interest rates and thought that this would slow down inflation. In fact, interest rates were not high enough. Relative to financial markets, the Fed was following rather than leading. It was reflecting the views of Wall Street. Finally, in the early 1980s, the “bond market vigilantes” took over, and we had high real interest rates, high unemployment, and a slowdown in inflation. Just to be clear, I am giving the credit for high interest rates to the bond market vigilantes, not to Paul Volcker.

Macroeconomics Without P

Posted on by

Scott Sumner writes,

I frequently argue that inflation is a highly misleading variable, and should be dropped from macroeconomic analysis. To replace it, we’d be better off looking at variables such as NGDP growth and nominal hourly wage rates.

If we cannot measure P tolerably well, then it sort of spoils the fun about talking about the real wage, the real money supply, or real GDP. I think that the consequence is that we shift all of the focus to:

— the average nominal wage rate, W
–total employment (or hours worked), N

The product of the two, WN, becomes aggregate demand (with NGDP an approximate indicator).

Aggregate supply determines the division between the two. For example suppose we have sticky nominal wages, with W depending on W* (what workers expected W to be, based on recent trends in nominal wages) and N (as employment gets closer to full employment, workers start to expect higher wages). We could have N affect the shape as well as the level of supply curve. That is, the effect of a change in N on wages could be low in a recession and higher near full employment.

From the Fred database, I have downloaded total compensation from the national income accounts (WN, in effect). And I downloaded total payroll employment from the establishment survey (N, in effect). The ratio of the two gives W. Some recent data on the percent change of these numbers.

Year WN percent change W percent change
2008 2.3 2.9
2009 -3.6 0.8
2010 2.3 3.1
2011 3.9 2.7
2012 4.0 2.3

Over the last five years, the median value for the percent change in nominal compensation has been only 2.3 percent. during the entire Great Moderation (1986-2007), there was only one year where nominal compensation grew by less than 2.3 percent (it was 1.6 percent in 2002). The median during the Great Moderation was 5.2 percent. Using this as a measure of aggregate demand shows weakness. Of course, any product involving N would show weakness, so don’t get too excited, folks.

The Phillips-Curve half of the story does not go as smoothly. Perhaps the 2009-2010 pattern is a fluke, and you should just average those two numbers? In any event, we had higher wage growth throughout most of the Great Moderation, but not all of it. From 1993-1996, annual wage growth was 2.1 percent, 1.8 percent, 2.1 percent, and 3.1 percent, respectively. What caused that episode of sluggish wage growth? In that case, it was not weak aggregate demand.

Market Monetarists Jump the Shark

Posted on by

Scott Sumner writes,

In America mortgage debt is commonly structured so that monthly payments stay constant over 30 years. This means that during periods of high NGDP growth, when nominal interest rates are also high, monthly payments will start very high in real terms, and then fall rapidly in real terms. But your ability to qualify for a house depends on how large the initial nominal monthly payment is, relative to your current income.

Read the whole thing. The logic is this:

1. In the 1970s, house prices started rising, but they did not rise as much as during the recent bubble.

2. In the 1970s, because mortgage rates were high, even though real interest rates were low it was hard to get mortgage credit. That is what choked off the bubble. The same thing did not happen in the recent bubble.

3. High mortgage rates reflect loose monetary policy. Hence, the difference between the 1970s and the recent bubble is that this time monetary policy was tighter.

I agree that there is a “money illusion channel” between nominal interest rates and housing. Back in the 1970s, economists proposed price-level-adjusted mortgages (PLAMs) to get around this problem. If you want more background, go to MRUniversity and watch the first half-hour or so of videos from my housing course.

But….come on. The extension of the recent bubble compared to the 1970s came from the abandonment of standards for down payments. If you think that looser monetary policy would have choked off the recent housing bubble, you’ve jumped the shark.

Greenspan and the Housing Bubble

Posted on by

Scott Sumner writes,

I really don’t care whether money was about right during 2005-06, or slightly too easy. Either way it wasn’t at all unusual compared to earlier periods of our history. Indeed during most of my life policy was far more expansionary during cyclical expansions than 2002-06.

John Taylor and some libertarian/Austrian economists judget Alan Greenspan as guilty for greatly exacerbating the housing bubble by keeping interest rates too low for too long. Scott Sumner exonerates Greenspan. I do, too, although for a very different reason. Sumner’s argument is that nominal GDP was not so far out of line. That is a fair point.

I would say that I find the strength of the link that Taylor finds between the Fed Funds rate and the housing market to be implausibly strong. The interest rates faced by borrowers are determined in the bond market, and the Fed’s influence there tend to be weak.

The other argument comes from the left, where it is suggested that Greenspan’s benign view of the markets blinded him to the excesses in credit creation that were fueling the bubble. In hindsight, this argument is compelling. Knowing what we know now, we can say that the Fed should have questioned the AAA ratings of securities backed by sub-prime loans, stress-tested banks on their exposure to a decline in house prices, and yelled “Danger!” about the collapse of credit standards at Freddie Mac and Fannie Mae. However, back when it mattered, in 2005 and 2006, not even the Bakers and the Shillers and the Krugmans who were talking about a housing bubble were recommending those actions.