Arnold Kling  

The Minsky-Jones Forecast Looks Good

A Certain Misperception... Outsider Academic Journals...

Two months ago, I wrote,

The standard forecast is, "employment has been less than what we expected, so let's assume it will continue to be less than expected." My forecast is that the recalculation is just getting started, and we will probably see employment growth of 200,000 jobs a month, or perhaps more, for the next year. Note that I make no bets about tomorrow's numbers.

That was before the release of figures for February. Today, we had the figures released for April, and the Bureau of Labor Statistics reported,

Since December, nonfarm payroll employment has expanded by 573,000, with 483,000 jobs added in the private sector. The vast majority of job growth occurred during the last 2 months.

In a Garett Jones economy, hiring is discretionary. It is not tied automatically to output. Instead, it is a form of investment. In a Minsky economy, when the economy is in a "hedge finance" stage, businesses are reluctant to borrow and instead finance investment out of profits. The profits started to come back a few quarters ago. Now, hiring is starting to come back.

Comments and Sharing

CATEGORIES: Macroeconomics

COMMENTS (10 to date)
Svend writes:

oh god damn. Be critical.
290K of which census was 66k and Birth Death was 188k. Hurray -the economy added a real 36k in jobs in April.

Doc Merlin writes:

The unemployment rate also rose.

OneEyedMan writes:

What are the seminal papers or works to read to understand the Minsky and Garett Jones economies?

fundamentalist writes:

JS Mills: The demand for commodities [ie, consumer goods] is not the demand for labor. Since Keynes, most economists have said Mills was wrong. The stats prove him right. Profits must revive before businesses start hiring because capitalists buy labor, not consumers.

Also, I think Hayek's Ricardo Effect kicks in here: the greatest unemployment is in capital goods makers, not makers of consumer goods (except durable ones like housing). When profits for consumer goods makers get squeezed by falling prices, they start buying labor-saving equipment (capital goods). That boosts profits for capital goods makers who begin hiring more workers.

The unemployment rate goes up at times like this because it is a survey of people looking for jobs. When people become more optimistic, more people start looking again and the rate goes up.

Doc Merlin writes:

"The unemployment rate goes up at times like this because it is a survey of people looking for jobs. When people become more optimistic, more people start looking again and the rate goes up."

Hrm, that makes sense, thanks fundamentalist.

Justin writes:


That 188k of net birth/death jobs aren't seasonally adjusted, they are raw numbers.

April is typically a month of heavy job creation - the BLS shows that the total number of nonfarm payroll jobs rose from 128,961,000 to 130,119,000, or 1.158 million jobs. However, most of those jobs are seasonal and there was a 290k increase in the trend. Another way to look at it is that only 16% of April's job creation was assumed by the birth/death model, or about 46,000 jobs if applied to the 290k seasonally adjusted payrolls number.

Given that the economy has been growing at a 3.7% pace over the past 9 months, I don't think it's crazy to expect some job creation occurring that isn't captured by the establishment survey. For that matter, the household survey suggests that over 1.6 million jobs have been created so far in 2010.

Also, the birth/death model generally does a pretty good job of estimating job creation that isn't captured by the establishment survey, with a key weakness being that it misses inflection points in the economy. It isn't perfect to be sure, but the model certainly seems to be unfairly maligned on the blogsophere.

Ella writes:

The economy hasn't been growing at 3.7%. GDP calculations have grown at 3.7% for three reasons ALONE:

1. Government spending
2. Inventory depletion
3. Reduced imports

And, although they don't count it in the rolling stats, consumer inflation. Green shoots!

I'm a bear, and I simply don't trust the government numbers because they simply leave out people they don't want to count as unemployed (self-employed people... people who've run out of benefits....) and they jigger the GDP numbers to make stuff look good - when actual buying doesn't bear it out.

Justin writes:

The economy has been growing at 3.7%, as best as we can measure it.

1) Government purchases of consumption and investment goods have actually been shrinking for the past 9 months at a -0.2% annual pace, per the BEA. Government has been a drag on growth (although isn't it always?) This is because a lot of the federal deficit is related to lower tax revenue, and state/local cuts are offsetting higher federal expenditures.

2) Inventory depletion (which leads to inventory building and higher production) is real growth. If an economy produces 80 units of goods in one period, with 85 units worth of sales, reducing inventories by 5, and then the next period produces 88 units of goods to restock inventory by 3 units, total output really did rise 10%. The rate of restocking isn't particularly fast yet either, given the depletion - it will probably continue to increase.

3) Reduced imports don't add to growth. This is what happens. You have an economy which consumes/invests 70 units, exports 10 units and imports 20 units. In your GDP calculation, you add up all consumption, add exports, and subtract imports, to get total production of 60 units. If imports fall to 15 units and consumption or exports also fall by 5 units, then the total effect on GDP is zero, because the economy is still producing 60 units. However, if imports fall to 15 and consumption/exports manage to stay unchanged, then production expanded by 5 units. It isn't that 'imports fell which adds to growth', it is that production rose, and our method of calculating production doesn't allow us to include imports so we subtract them out, and oh by the way imports are down.

Yes, there is consumer inflation but it also gets stripped out as best we can measure it (faulty as it is).

While the government does have very particular definitions for different measures of unemployment, consider its measurement of employment at the household level. Over the past four months, employment as reported by households has expanded by more than 1.6 million. Since these numbers survey households directly, they capture the self-employed (and these surveys are more optimistic than surveys of companies to generate payroll numbers).

What evidence is there that the government is deliberately manipulating statistics to make things look good? I've seen ShadowStats but those numbers are bogus - if those inflation numbers were true my standard of living would have fallen in half over the past decade, and not risen. I'm not saying government statistics are perfect (I don't think anyone can say what the 'true' rate of inflation is - they can just provide different measurements of it), and production/employment are still below their peaks, but I think Arnold is right that the profits have come back and now jobs are starting to come back. The recalculation appears to be yielding some fruit.

Kevin writes:

Justin you're doing God's work, so to speak. But Ivandjiiski's ZH acolytes aren't really down with that - like their masters they manage, laughably, to miss even Palahniuk's simple point.

I believe the dynamic Minsky describes is under way, but I question the speed at which a recalculation can happen when it's dependent on equity financing.

MernaMoose writes:

Just for what it's worth, in the S&T (science and technology) universe that I live in, hiring is slowly coming back. Though it's not clear to me how and why.

At times it seems to be driven more by government spending than anything else. But my perception may be warped by the market segments I work in.

Curiosity: the hiring I do see, tends to be on the high end of the job skills scale. It's not run-of-the-mill, down-in-the-trenches design engineers that I'm seeing demand increases for, but rather it's for the high powered PhD types.

Another for-what-it's-worth observation: over the past five years, there's no getting around the fact that computers and software advances have made us S&T types more productive. My personal feeling is that people are starting invest again in this same vein, because it's pretty clear that another round of productivity improves is there to be had -- with investment.

Comments for this entry have been closed
Return to top