Paul Engel. Re: "The big manufacturing push continues. "
The WSJ carried an interesting column yesterday by Steve Liesman, which talked about a couple of interesting papers linking microchip speed to productivity gains and overall economic growth. It shows how important Intel's cap spending is to the overall economy.
HL
The WSJ states:
A lot of what the New Economy created has gone poof: the technology bubble, the dot-com upstarts, fat corporate profits and outsized economic growth. But the thing that started it all -- a revival in productivity growth -- has stubbornly hung around.
During the first two full quarters of the official recession, productivity, or output for hours worked, has averaged a solid 1.8% growth, compared with an average of 0.14% in the first two quarters of the previous nine recessions. Productivity growth in the fourth quarter also appears to have been strong.
On Monday, at a conference at the Atlanta Federal Reserve Bank, two leading experts in the field of productivity will present a research paper arguing that such strength is likely to continue for a while. The work by Dale Jorgenson of Harvard University and Kevin Stiroh of the New York Federal Reserve Bank says that the likely scenario for productivity growth over the next decade remains a robust 2.24% annually. That's just a tad lower than the average 2.36% that helped the economy surge from 1995 to 2000.
"The U.S. productivity revival remains largely intact," the two say in their paper, which was written and researched along with Mun S. Ho, of the think-tank Resources for the Future in Washington.
If they're right, there are important consequences for wages, interest rates and growth. When productivity rises, employers can pay higher wages because they are producing more with less. The Federal Reserve doesn't have to fret as much about inflation because output grows without straining resources. Ultimately, that results in a bigger economic pie.
Behind their optimism, the authors see little change in the forces they believe created strong productivity-growth rates in the latter half of the 1990s; a fall in prices, they argue, drove companies to invest in productivity-enhancing technology goods, and those price declines are continuing. The price declines themselves stemmed from an accelerated pace of technological innovation, principally in the speed with which more powerful microchips were brought to market; whether we get more of the same is thus a key question.
To expect strong productivity growth to continue seems an unlikely forecast amid one of the largest capital-spending slumps on record. But in an interview, Mr. Jorgenson explained that the forecast doesn't require a return to the outsized spending levels of the late 1990s. Yet he does see a substantial rebound in the making, driven by technology prices that are now falling even faster because of the recession. "These price trends are going to move people in the same direction that they did in the past," he said, which is toward greater capital spending on information technology.
The paper puts the long-run, noninflationary growth rate of the economy -- the so-called speed limit -- at 3.34% over the next decade, about a percentage point higher than what economists believed was possible before the productivity surge. However, the figure represents a sharp reduction from the average annual growth rate of 4.6% from 1995 through 2000, with the main difference being a reduction in the growth of hours worked.
To be sure, some skeptical economists have questioned whether technology's contribution has been overstated and the extent to which productivity gains have been concentrated among the chip makers and computer and telecommunications producers, and not the users. And the Jorgenson study offers a wide range of potential productivity-growth rates, from 1.33% to 2.98% annually.
What will most likely determine the outcome, they say, is whether chip makers can develop faster semiconductors on a two-year or on a three-year cycle. The authors' likely scenario assumes a middle ground.
A very significant variable in the forecast is how much companies invest in these new technologies. In a paper, Jason G. Cummins of the Federal Reserve Board in Washington and Giovanni L. Violante of University College London look at the so-called technology gap in the economy: how much more productive the best machines are compared with the average machine.
Their findings are startling. In 1975, the productivity of the average technology used on the factory floor lagged the best by just 15%. In 2000, the figure jumped to 40%. The gap actually widened in the 1990s, to 40% from 35%, despite the capital-spending boom. What this shows, Mr. Cummins said in an interview, is that "the frontier is moving out so quickly, the average practice has continued to fall behind."
It also suggests that you "have to do a ton of investment to keep pace with the frontier" and that such investment pays dividends.
Both papers support the argument that it is reasonable to project into the future the recent productivity gains of the past. In other words, despite the current recession, a big part of the New Economy story remains unchanged.
It's a difficult concept to take with the Nasdaq off 59% from its high. But Robert DiClemente, chief U.S. economist at Citigroup Inc., says it's not so confusing. He says many people have simply mixed up the underlying trend of productivity growth with the fact that investors overvalued that trend by bidding up tech stocks or by building too many fiber-optic networks.
The same danger exists going forward, however: figuring out how to profit from productivity growth can be about as hard as forecasting it.
Copyright The Wall St. Journal |
