...and finally, argument by unfounded assertion

It bothers a lot of people that food and energy are excluded from the inflation measure that is used by the Fed as a guide to policy.

It bothers Jayhawk:

So how did the “Trump tax cut” cause the decline in new home sales? Well, the tax cut caused runaway inflation, which we didn’t see but Dean Baker and the Federal Reserve did, when it soared to dizzying heights and got all the way up to 1.87% last year.

Never mind that the price of gasoline went up 74%, it’s all in how you measure things. Dean Baker and the Fed only measure the prices of things that people don’t buy. Food, energy and housing are not included, thus 1.87% inflation.

Yeah, it bothered me too. A lot. But I read something that rang a bell, and now it doesn't bother me at all. At least for now.

Randal K. Quarles:

Traditionally, as taught in Econ 101, inflation provides a signal on whether the economy is operating above or below its potential level. If inflation moves up in a sustained manner, not just because of temporary shocks, then the economy is likely operating above its productive capacity, as firms have the leeway to raise prices given the strength of demand.

Three things about that statement. First,

inflation provides a signal on whether the economy is operating above or below its potential

Yeah, okay, that's mainstream econ. The dumbed-down "Econ 101" version maybe, but mainstream. I have just one problem with it, which I'm saving for last.

Second,

If inflation moves up in a sustained manner, not just because of temporary shocks, then the economy is likely operating above its productive capacity...

Just the way Quarles said it, I get it: if inflation moves in a sustained manner, not because of temporary shocks. The Fed is looking for evidence of sustained inflation, not for evidence of brief disturbances. They're looking at the trend, not the jiggies. It makes good sense.

I think that's probably a good way to set policy. But I don't think it's a particularly good way to measure the inflation that affects us as consumers, or whatever we call ourselves. I'm still with Jayhawk on that. If the price of gasoline went up 74% because of some "shock", we still had to pay the higher price. The cost of living includes it all, you know? Even food and gasoline.

So I figured I'd make a graph comparing the PCE Price Index with, and without, those "shock" categories. And then it got interesting:

Graph #1: PCE Price Index including (red) and omitting (blue) Food and Energy Costs

Moving together in 2012 and 2013, the two datasets separated in 2014. And since that time, the one that includes food and energy has been lower. So, by this measure anyway, leaving out food and energy shocks makes the price index higher. I can't tell you why, but that's what the graph shows.

I need another look. A second graph. This time, showing the index that omits food and energy, relative to the index that includes them:

Graph #2: Excluding Food and Energy, relative to Including Food and Energy

If "shocks" to food and energy prices were pushing the rate of inflation up, then the index that doesn't count them would be lower than the one that does. But the index that doesn't count the shocks is higher, since 2014.

So the Fed is using the higher measure of inflation as a guide to policy. Not the lower measure. And you really cannot say what Jayhawk seems to imply, that they're excluding food and energy to reduce the calculated rate of inflation.


"...and finally, argument by unfounded assertion"

Getting back now to what Randal Quarles said, there is one more point I have to make.

If inflation moves up in a sustained manner, not just because of temporary shocks, then the economy is likely operating above its productive capacity, as firms have the leeway to raise prices given the strength of demand.

If there is a sustained increase in inflation, Quarles says, it is because of the strength of demand. In other words, all inflation is demand-pull.

The purpose of prediction

The purpose of making prediction based on theory is that it is a test of the theory. I read this, somewhere. Anyway, it means that if events turn out as predicted, the theory is not rejected and may deserve increased attention.

Back in the Spring of 2016 I said the economy had hit bottom and things were picking up. I predicted vigor. I also said it would be a couple years before the vigor was noticeable, and that it would last for several years. My prediction was based on the theory that when financial cost in the economy is low but rising, growth is good; and when financial cost is high it consumes money that otherwise might have gone into growth.

Anyway, this morning I find Randall Quarles in a speech of 18 October 2018, saying:

... in February, I characterized the U.S. economy as being in a "good spot" and asked if the economy had reached a positive turning point following an extended period of post-crisis slow growth. I argued that while it might be too soon to call a turning point, there was a definite possibility of an upside surprise.

So now that we are fairly deep into 2018, where do we stand overall? My view has not changed all that much from February. While many other forecasters had to revise up their forecasts over the course of the year, my own outlook is basically unchanged, because the economy is evolving essentially as I expected at the outset of the year. The economy remains in a good spot...

The predicted improvement became noticeable to Randal Quarles early in 2018. Right on schedule.

Quarles also says

How long can this strong growth be sustained? ...

... I see many reasons to be optimistic about the growth of the potential capacity of the economy over the next few years. In part, my optimism is rooted in the view that many of the factors that have been weighing on potential growth since the financial crisis could be lifting. So, have we reached the turning point? While I believe the issue remains unresolved, the recent evidence is encouraging.

The strong growth will last as long as people are willing to expand their debt obligations.

// See also: my Vigor page.

How to reduce our trade deficit

Keynes, from Chapter 23:

If the domestic rate of interest falls so low that the volume of investment is sufficiently stimulated to raise employment to a level which breaks through some of the critical points at which the wage-unit rises, the increase in the domestic level of costs will begin to react unfavourably on the balance of foreign trade, so that the effort to increase the latter will have overreached and defeated itself.

Paraphrasing: If unemployment falls below the NAIRU, rising wages will create inflation. Rising prices will make domestic products less competitive on world markets, and will tend to push the balance of trade toward deficit.

In other words, trade deficits may be caused by high domestic costs.

You can accept this conclusion, or reject it. I accept it.

If you accept it, you can accept or reject the view that the rising wage is the only cause of high domestic costs. I reject it.

If you reject the view that the rising wage is the only cause of high domestic costs and the resulting trade imbalance, then you are free to consider other explanations.

In my view, the cost of a large and growing financial sector is the prime mover that raised domestic costs in the post-WWII period.


Your President thinks the trade imbalance is best reduced by tariffs on imports.

I think the trade imbalance would be best reduced by adding cash to the economy while reducing policy incentives to borrow and, at the same time, implementing policy incentives to repay debt. By driving the cost of finance out of products, we can reduce costs and make our products competitive on world markets again.

The last chapter

The first sentence:

The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and incomes.

One interesting economist

Recommended skimming: Modernizing Monetary Policy Rules by James Bullard.


"Interesting": Finding new ways to look at accepted ideas, resulting in new ideas. I'm thinking of Bullard's earthquake and his 2½-year forecast horizon in addition to the "modernizing" piece.


// Two brief thoughts on Bullard's Modernizing:

1. People look for all sorts of ways to get the inflation number down. Not necessarily to get inflation down, but to get the number down. Bullard subtracts three tenths of a point from the CPI inflation rate, because CPI inflation is higher than PCE inflation.

Is it a reasonable adjustment? Maybe. Probably. But it is just another way to bring the inflation number down.

2. Bullard says

the general level of short-term real interest rates has been trending lower for three decades

He does not talk about why rates have been falling for an extended period. He just goes with the fact, and makes use of it in his calculations.

Granted, the reasons that rates have been falling are not relevant to his topic. I'm just saying he shouldn't have moved on to this topic until the reasons rates have been falling have been thoughtfully and thoroughly examined.

So now of course I will have to search through his works, looking for his thoughts on the reasons rates have been trending lower for three decades. Not today. But I'm thinking that when I do, I'll find Bullard generally accepting the view of John C. Williams and others, that demographics is the main reason for the long-term decline in rates.

The path of interest rates has not been thoughtfully and thoroughly examined unless the discussion includes debt, private debt, its effects on the economy, and the unintended consequences of policy which created such high levels of debt.

Surprised me

From Robert T. McGee's Applied Financial Macroeconomics and Investment Strategy, pages 36 and 37:

As mentioned in chapter 1, over long periods of time job growth seems to depend mainly on how many people are available to work, that is, the growth of the labor force. That's because the trend growth rate of the economy is determined by the labor-supply growth rate and the growth in its productivity. Interestingly, low labor-force growth seems to put pressure for stronger productivity growth. This seems to have been the case in the 1950s, when the low-birth cohort from the Great Depression came of age to work. Despite a slow-growing workforce, productivity was higher, and the economy grew at a respectable rate. Conversely, when the huge babyboom generation was coming of age in the 1970s, productivity growth dropped significantly, suggesting cheaper, abundant labor was substituted for relatively more dear capital.

Reminds me of what Menzie Chinn said (January 2017) about Donald Trump's 3½-to-4% growth target:

In order to hit the lower bound of the Trump target for 2017-2020, either contributions from labor force growth, or labor productivity, or combination thereof, must accelerate by 1.8 percentage points.

An increase in output requires either an increase in hours worked or an increase in output per hour, or both. Chinn called this "growth accounting". Here's the graph he showed:



Okay, so Chinn and McGee agree on the growth accounting. But that's not why I'm quoting McGee. I'm quoting McGee because he says low labor-force growth seems to create pressure for stronger productivity growth. He says it's interesting, and I agree; but I think he means to suggest it is surprising. I certainly found it surprising.


I accessed the FRED index series Nonfarm Business Sector: Real Output Per Hour of All Persons, changed it from quarterly to annual data, and imported it into Excel. Figured "percent change from year ago" values, and took averages for time periods to match the graph Chinn showed. Except I stopped at 2017, the last year for which I had data.

Then I got the FRED series Civilian Labor Force and set it up the same way. I put the two together on a graph, and copied the 2017 value out to 2027 so my graph ends when Chinn's graph ends.

Do productivity growth and labor force growth tend to move in opposite directions?

Well, yes they do. Except during the special circumstances of 2007-2008, all of the changes are in opposite directions.

They do move in opposite directions. From the annual data, I would never have guessed:

I've been looking at this all wrong

Not: Interest Paid relative to GDP...
But: Interest Paid relative to the money we use to pay for things.

Graph #1