May 29, 2003

Our Current Semi-Recession

The Wall Street Journal's John Hilsenrath thinks about the fact that the rapid productivity growth of the "new economy" means that output has to grow faster than 3.5% per year for unemployment to fall. That the productive capacity of the American economy is rising so rapidly is a powerful reason for more stimulative short-run policies--and is the thing that makes the Bush administration's failure to propose a significant, real stimulus package a considerable shame.

WSJ.com - This Recovery Feels Like Recession: Economy Expands, Payrolls Shrink: ...Payrolls in the electronics sector, and for producers of industrial equipment, have declined for 28 straight months. In communications, payrolls have fallen for 24 months. In the securities and airline industries, they have fallen in 16 of the past 24 months.

In some ways, this is the downside of a productivity boom that created much optimism about the economy during the 1990s. Productivity growth means that companies are squeezing more output from existing workers. Over the long run, most economists agree productivity growth is good for workers, because it tends to lead to higher wages. But in the short run, it is creating a problem. Worker productivity has been growing faster than the overall economy. That has allowed corporate executives to meet small increases in demand while still eliminating jobs.

"You end up with a jobless recovery," says Jared Bernstein, a labor economist with the Economic Policy Institute, a left-leaning think tank in Washington. "It is indistinguishable from recession for many working families."

A common definition for a recession is two consecutive quarters of contracting gross domestic product. The nation's GDP -- the broadest measure of economic output -- has expanded at an average annual rate of 2.7% since the fourth quarter of 2001. During the same period, the productivity of the nation's work force -- which is defined as its output per hour of work -- has expanded at a much faster rate of 4.2%. While worker productivity often increases in the early stages of a recovery, this time the mismatch between productivity and overall economic growth is unprecedented.

At the beginning of eight recoveries between 1948 and 1982, GDP grew faster than productivity. In those cases, companies had to add workers to meet demand for their goods and services. During the recovery of 1991, productivity grew slightly faster than output in the early stages, but the difference wasn't as stark as it is now.

"If you want people to have jobs, your demand-side growth has to be much stronger," says Harry Holzer, a labor economist at Georgetown University. The nation would need a 3.5% growth rate in GDP for the unemployment rate not to get worse, he says, but "3.5% is looking optimistic for this year. This might be quite a protracted downturn."

Posted by DeLong at May 29, 2003 07:51 AM | TrackBack

Comments

By "considerable shame" you mean "source of wholly unnecessary pain for millions of americans and their families", right?

Posted by: FMguru on May 29, 2003 10:41 AM

Isn't it more likely that the loss of jobs is a result of companies finding ways to flush deadwood? In rich times the hard decisions are difficult to make and getting rid of speculative workgroups that never find a way to be productive or misplaced employees is slow. But when cost cutting is essential, those unproductive employees are laid off.

There is no productivity increase.

The apparent increase is a side effect of the definition of average productivity. It's output per worker hour. But if some workers are not even intended to be productive, the formula is misleading. Get rid of those jobs which do not contribute to productivity and the output will stay the same while the apparent productivity increases.

That has been the pattern in telecom and technology sectors.

Employment will increase when companies can't find anymore deadwood to throw overboard. Or when the rate of new hiring for new projects exceeds the rate of deadwood disposal.

It might be a good thing to raise interest rates to help flush out unproductive divisions staying afloat only by borrowing cheaply. Probably we should have raised them back in 2001-2002; tech employment would be a lot better today.

Posted by: Newt on May 29, 2003 10:41 AM

Isn't it more likely that the loss of jobs is a result of companies finding ways to flush deadwood? In rich times the hard decisions are difficult to make and getting rid of speculative workgroups that never find a way to be productive or misplaced employees is slow. But when cost cutting is essential, those unproductive employees are laid off.

There is no productivity increase.

The apparent increase is a side effect of the definition of average productivity. It's output per worker hour. But if some workers are not even intended to be productive, the formula is misleading. Get rid of those jobs which do not contribute to productivity and the output will stay the same while the apparent productivity increases.

That has been the pattern in telecom and technology sectors.

Employment will increase when companies can't find anymore deadwood to throw overboard. Or when the rate of new hiring for new projects exceeds the rate of deadwood disposal.

It might be a good thing to raise interest rates to help flush out unproductive divisions staying afloat only by borrowing cheaply. Probably we should have raised them back in 2001-2002; tech employment would be a lot better today.

Posted by: Newt on May 29, 2003 10:42 AM

"It might be a good thing to raise interest rates"

I suspect they're afraid of bursting what may be a housing bubble.

Posted by: Steve Antler on May 29, 2003 11:21 AM

Hey ho, hey ho. "Flush the deadwood." The hell with working women and men, flush em. Love ya spirit.

Posted by: lise on May 29, 2003 11:50 AM

There may be a grain of truth truth in Newt's comment, but it misses the point: the recovery from this recession is anomalous.

As to Newt's point itself, productivity growth was also quite high before the recession, and fell sharply during the first half of 2001 as these "unproductive" people were being laid off. So if it is true that average productivity increases because of eliminating deadwood, this likely occurs at the margins.

Second, this does nothing to explain why the gap between productivity growth and GDP growth should be historically astronomical. I assume that proximity to the liquidity trap and the Bush administration's unwillingess to devise a sensible stimulus plan are among the chief culprits.

Posted by: Henry on May 29, 2003 12:32 PM

There may be a grain of truth to Newt's statement, then again, there may not. He pretends to have dipped his hand into some mine of economic reality and pulled out the gem of productivity for examination. He knows there is no productivity increase. How?

The WSJ offers another view on productivity on C1 today. That is that there was indeed a rise in productivity, that it was based on technology and that the slow pace of capital investment puts productivity gains at risk.

Posted by: K Harris on May 29, 2003 01:00 PM

I'm afraid I don't see the downside to the increased productivity growth (well, I do see a downside as far as the environment, actually, but no employment-wise downside). After all, suppose productivity growth had NOT accelerated, but history was otherwise largely the same. So, what makes Hilsenrath think that we'd be having the same average 2.7% growth, combined with 1% productivity growth to make a nice, job-growthy recovery, rather than our current high-productivity growth, job-losing recovery? After all, the Fed's policies would have been different prior to the recession, and the general rate of growth would have been slower throughout the 1990s and early 2000s. Why wouldn't we have, say, no GDP growth at all combined with 1% productivity growth, and a jobless NON-recovery?

Posted by: Julian Elson on May 29, 2003 01:02 PM

Bush fiscal policy has a neat solution for this issue. If aggregate demand growth will be only 3% per year, make sure national savings will be so low that long-term labor productivity growth is low enough that total real GDP growth at full employment will also be 3% per year. It's a terrible idea for the well being of the citizens but the employment stats will look OK.

Posted by: Hal McClure on May 29, 2003 01:49 PM

There really does not seem to be a slowing of investment in technology or a slowing of productivity growth. Technology spending has been running at a decent clip, but prices for performance have been coming down, so we can buy more and more performance with a fairly constant technology budget.

I expect productivity growth to stay high enough to require about 4% GDP growth for a sustained period to begin significant job creation. I wish I were more optimistic about a resurgence of demand driving GDP growth and investment.

Posted by: anne on May 29, 2003 01:58 PM

I remain unclear about cause and effect. Is worker productivity a real and measurable quantity, or is it just the result of dividing GDP by employment? It seems to me (without any facts, but that's what comment threads are for) that we are seeing significant price pressures across the board, resulting in layoffs and increased demands being made on the remaining employees. But on the consumer side, low prices are resulting in increased consumer spending, resulting in higher GDP. I guess my question is how much harder business can continue to squeeze employees to continue generating such striking productivity growth.

any thoughts?

Posted by: FDL on May 29, 2003 02:08 PM

What I would like to see is someone's idea of a "real" stimulus package, and the subsequent path back to a balanced budget. Are we talking about a one-year deficit of $800 billion, and then an immediate lurch back to a balanced budget? Or are we talking about an initial deficit of $650 billion, followed by $600 billion, then $550 billion, ...

I really would like to know what we're talking about here.

Posted by: Arnold Kling on May 29, 2003 02:31 PM

Newt-

All innovation begins as speculation and it is not easy to know the risk up front. So I don't know how you can be so sure that cutbacks in those areas involve deadwood and not future gains in productivity. I would agree though that a lot of workers were involved in roll-outs of new technology and not in production, and a lot of the continuation of productivity involves the cutting of the "reorganizers" and leaving just the reorganized. So if that's true then this effect will run its course.

Also, hard decisions are always being made in good and bad times, because competition does not go away, and efficient and paranoid companies are always driving out companies which fall asleep at the wheel or are inherently less efficient. A worker who is deadwood at one company but could be doing something more productive elsewhere should be offered more money to switch. Job hopping happens more so during good times. Why is it better if the economy is so slack that the worker instead of switching jobs just loses theirs? Instead of trading some output for more output you trade some output for none. High wages and competition for labor are just as good of a reason to check yourself for efficiency as are a drop in revenue, although any company with competent management should always be looking to increase profitablity.

Don't fall for this liquidationist stuff.... suffering to better yourself is a good description of investment, but not of recessions. Raising interest rates as you suggest to purge the unproductive would also remove the productive, because it can't single out the decisions which will turn out to be incorrect in the future and leave the decisions which will be fruitful. If we knew which was which then there would be no unproductive groups to begin with. And don't forget that the goal is to maximize our well being and not productivity. If productivity was the goal in itself we could just go ahead and fire the 50% least efficient workers.

Posted by: snsterling on May 29, 2003 02:45 PM

The long run not only raises wage levels, it also moves the starting point for future changes. So the base level of growth needed to absorb all unemployment also rises, and the rate of increase of productIVITY goes up because growth is high... so there is scope for a vicious circle, much as in the logic of appendicitis or congestive heart failure (where the body's protective reactions also boost the underlying harm process).

And, of course, in situations of change the equilibrium is not so much an attained condition as a reference point, the way a centre of gravity need not be inside the parts of a solid structure but between them. What counts there is the steady state behaviour, if even that is attained (just add cyclic factors and shocks, if not); it's quite possible for a steady growth process to have a constant proportion of lag from levels yielding full employment - or the other way round, to be used to keep employment above the NAIRU if you can only cut it.

Posted by: P.M.Lawrence on May 29, 2003 04:28 PM

Arnold Kling, the Democrats have proposed a real stimulus package that is on Congressman Spratt's budget committee website. The Democrat proposal has received almost zero press coverage and is not taken seriously because the GOP controls Congress and the WH.

Basically, the package contains tax breaks for low income wage earners, revenue to the states for unfunded mandates and a few other goodies. Their total package was around $120 Billion after offsets in subsequent years. The Democrat package has some non-stimulus items such as a permanent 10% bracket. If the non-stimulus items in the Democratic package were frontloaded as rebates rather than permanent tax rate cuts, then a revised package would create about 1 million new jobs for about $24 billion.

That works otu to about $24,000 per job, a bargain. It could be done, but it demands putting aside politics and ideology and putting the money where it will do the most good.

Mr. Bush is trying to deficit spend the economy into recovery. If he runs enough deficit, eventually it will work but at what cost?

Posted by: bakho on May 29, 2003 06:57 PM
Post a comment