Message-ID: <19682438.1075843091500.JavaMail.evans@thyme> Date: Mon, 1 Nov 1999 11:14:00 -0800 (PST) From: levine@haas.berkeley.edu To: e201b-1@haas.berkeley.edu, e201b-2@haas.berkeley.edu Subject: A recent BusinessWeek article on measuring GDP and productivity Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-From: "David I. Levine" X-To: e201b-1@haas.berkeley.edu, e201b-2@haas.berkeley.edu X-cc: X-bcc: X-Folder: \Jeff_Dasovich_Dec2000\Notes Folders\Mba--macroeconomics X-Origin: DASOVICH-J X-FileName: jdasovic.nsf Srikanth Hari sent the following BusinessWeek ANALYSIS & COMMENTARY I thought others might want to read... Software: A New Calculus for the New Economy How reclassifying it as an investment boosts growth The U.S. truly has seen the birth of a New Economy over the past several years. That's one way to read the results of a comprehensive revision of historical data on the gross domestic product released on Oct. 28 by the Commerce Dept.'s Bureau of Economic Analysis. The most stunning data in the report are about the acceleration of productivity in the 1990s. Official revisions of productivity data, incorporating the latest output figures from Commerce, won't be released by the Labor Dept. until Nov. 12. But a BUSINESS WEEK analysis of the new data from the Commerce Dept. shows that nonfarm business productivity growth in this decade will likely be revised upward, to roughly 2% a year, from 1.4%. Productivity growth will be boosted for the 1980s as well, but not by as much (chart). What's even more striking is the upward revision in productivity growth for the last three years, from 1.9% to a stunning 2.6% rate. That explains how the U.S. has been able to grow at a 4.2% rate over that stretch without creating inflationary pressures. KEEPS ON TICKING. While the rate of increase in economic output in the 1990s isn't as impressive as the growth in productivity, it is not as low as the old statistics indicated. From 1990 through 1998, the Commerce Dept. statisticians say, the economy grew at an annual rate of 3.1%--about even with the 1980s growth rate of 3.2%. And, the report confirms what everyone senses: Growth has accelerated recently. Commerce revised the economy's growth rate for 1996-98 upward to an average of 4.2% from a previously reported 3.8%. By comparison, annual average GDP growth for the entire 40-year span covered by the revisions was pushed upward to 3.4% from 3.2%. The implication of the revisions: The U.S. economy may now be able to sustain a higher rate of growth without inflation than was previously believed. The Commerce Dept. now says that the GDP last year was $8.76 trillion, which is about $250 billion--or 3%--higher than previously believed. A new calculus for software investments accounted for about two-thirds of the upward revision in GDP. And since software sales are growing far faster than the economy as a whole, adding them into the GDP raises the economy's official growth rate--and will likely continue doing so for years to come. Indeed, the biggest factor in the Commerce Dept.'s revisions is the new way of accounting for software, the poster product of the New Economy. The government now treats software purchases by business and government as investments, which are part of GDP. Until now, the only software sales that counted toward GDP were sales to consumers and the sale of software that came bundled with computers and was included in the computer's purchase price. Now all software sales are factored into the GDP figures. The new treatment of software isn't just some statistical trick. Until now, software spending by business and government was considered an expense, and expenses are left out of GDP to avoid double-counting. For instance, it would be double-counting to list the steel that goes into a car as part of GDP and count the car, too. But software is not consumed in the production of new goods and services. And the government is acknowledging now that software shouldn't be considered an expense such as steel or stationery. It's being reclassified as an investment, like a machine tool, in recognition that it has a useful life of at least a year. PIGGY-BANK BLOAT. By recognizing software as an investment instead of an expense, the Commerce Dept. also increases the national savings rate. Savings are the mirror image of investment. While the national savings rate was revised downward for 1959-73, it was revised upward for the years since. The biggest revision was for the most recent year covered: 1998. The government now says that total national savings were 18.8% of gross national product last year, a jump of 1.5 percentage points from the previous calculation. Of that increase, fully 90% was accounted for by the government's recognition of business and government software purchases as an investment. The Commerce Dept. also rejiggered the composition of the national savings accounts, raising personal savings while lowering government savings by an equal amount. That's because of a new way of accounting for government pension plans, contributions to which are now considered a form of personal rather than governmental savings. The reshuffling means that the personal savings rate is now calculated as 3.7% for 1998 instead of a previously reported 0.5%. The new methodology will undoubtedly turn the current personal savings rate from negative to positive. But since the government's savings fell by a corresponding amount, the reshuffling had no impact on the overall national savings rate. And the downward trend in personal savings is still in place. By the new measure, the personal savings rate declined from 10.9% in 1982 to 3.7% last year, instead of from 9% to 0.5% as had been previously reported. KINDER HINDSIGHT. The GDP revisions were wide-ranging and included hundreds of large and small tweaks. For instance, the government is now using a better way to measure the value of banking services such as automated teller machines and electronic funds transfers. That increases the sector's official output and productivity. Also, new inflation measures developed by the Labor Dept. are being incorporated all the way back to 1978, not just since 1995. One footnote: The government now says the recession of 1990-91 was milder than previously thought. From its peak in the second quarter of 1990 to its trough in the first quarter of 1991, GDP fell 1.8%. The old figure was a 2.7% drop. The Commerce Dept.'s rewrite of economic history doesn't mean that the Federal Reserve can afford to be any less vigilant about inflation. Regardless of the official growth rate, Fed Chairman Alan Greenspan has to keep his eye on the same warning signs: tight job markets, busy factories, rising commodity prices. The real significance is that government statistics are finally reflecting the reality of the high-growth, high-productivity, low-inflation New Economy. By Peter Coy in Washington My comments: The article is a bit unclear on whether software was ever counted. The key point is that the value of software not yet depreciated is a percent or 2of GDP. Thus, GDP is a few % larger than the old accounting system that expensed all software immediately. Consider your own desktop: What % of your salary is the market value of the software you have on the machine? (Not the original cost, as Word95 has a very low market cost, even if it was $500 4 years ago.) David I. Levine Associate professor Haas School of Business ph: 510/642-1697 University of California fax: 510/643-1420 Berkeley CA 94720-1900 email: levine@haas.berkeley.edu http://web.haas.berkeley.edu/www/levine/