The Economist on measuring investments in intangibles

Getting a grip on prosperity

Mar 2nd 2006
From The Economist print edition

What changes if intangible investment is measured properly?

ARGUING that America is ill served by its economic statistics has become fashionable. Stodgy old national-income accounts are said to do a poor job of measuring the modern “knowledge” economy. They are especially bad at picking up firms’ “intangible” investments, such as building brands or training staff. Measure this spending properly, the argument runs, and America’s true economic health would be revealed. Investment and output growth would be higher; the pesky external deficit would be lower. As Business Week put it recently, the “unmasked” American economy is “a lot stronger than you think”.

Really? It would certainly be convenient if America’s much-vaunted macroeconomic weaknesses (a falling saving rate, over-reliance on foreign borrowing to finance consumption) were mere statistical mirages. Unfortunately, things are not so simple. Although measuring intangibles does change the picture of America’s economy, it does not necessarily improve it.

Under today’s international system of national-income accounting, overall investment includes firms’ spending on fixed capital (such as factories or computers), additions to inventories, and expenditure on residential and commercial property. Spending on research or organisational improvements, in contrast, is not counted as investment but as spending on current inputs. Because national accounts are calculated on a value-added basis, it drops out of overall GDP. Granted, official statistics are evolving. Many countries now class expenditure on software as investment. Statistical agencies are also discussing the capitalisation of research and development spending. But other intangible investment goes uncounted.

The intangible mask

Few economists doubt that boosting a brand can be an investment, just as building a factory is: conceptually, investment is simply spending today that yields a stream of income in the future. The problem is practical. Intangible investment is often hard to identify, let alone value. How much of an advertising campaign is part of the cost of selling a product (and thus current spending) and how much is investment in a brand?

In recent research Carol Corrado and Dan Sichel, of the Federal Reserve, and Charles Hulten, of the University of Maryland, have tried to estimate the scale of intangible investments in America. The authors look at spending on all manner of things, from developing copyrights to tuition payments for job-related training. Often they are forced to make heroic assumptions, guessing, for instance, that managers spend 20% of their time on organisational innovation (an investment) and the rest running their firms. Given such guesswork, the results should not be taken too precisely. But they are striking nonetheless.

The authors find that the pace of intangible investment by American firms has risen sharply in recent decades: by the late 1990s, it was around $1 trillion a year, about the same as expenditure on traditional fixed assets. According to today’s official statistics, America’s investment rate has been pretty flat since the 1950s. Count intangibles, and it has been rising steadily (see chart). In the 1950s the inclusion of intangible investments would have raised GDP by 5%. In recent years it would have lifted it by 12%. Using yet more assumptions (this time about depreciation rates), the authors reckon that excluding intangibles may understate America’s capital stock by $3.6 trillion. Even if this is only roughly right, intangibles seem too big to ignore.

Thanks to higher investment, including intangibles pushes up productivity growth too–although it does not explain the acceleration of productivity after 1995. Capitalising intangible spending raises average productivity growth between 1973-95 by more than it does for the years since then.

A striking shift occurs between labour and capital. Official statistics say that workers’ share of national income has been fairly flat for decades. Capitalising intangible investment increases capital’s slice and reduces labour’s, with a particularly sharp drop after 1980. If intangibles are included, labour’s share today drops from 70% to 60%.

If investment is higher than today’s statistics suggest, so, automatically, is saving. With firms’ current spending lower and profits fatter, corporate saving (and so national saving) must be higher. It follows that the capitalisation of intangibles alone does not affect America’s external deficit, which is the difference between what the country saves and what it invests.

Not exactly, argue some knowledge-economy optimists. First, they claim, ignoring intangibles affects America’s external accounts directly. Were intangibles measured properly, the current-account deficit might be smaller, since America exports lots of managerial know-how and other intangible assets. Two Harvard economists recently labelled these and other uncounted exports “dark matter”. To most other economists, however, this argument is thin. Even if intangible exports were counted explicitly (rather than as part of firms’ profits on foreign operations), they would not alter the overall reliance on foreign funds.

The optimists’ second argument is that the reliance on foreign capital may be real, but it is less worrisome. Borrowing to boost skills or research is quite different from borrowing to fuel a consumption binge. Close examination of Ms Corrado’s, Mr Sichel’s and Mr Hulten’s estimates casts doubt on this too. Their numbers suggest that intangible investment slowed sharply after 2000, the year the stockmarket bubble burst–as America’s current-account deficit soared. Foreigners may have been funding intangible investment in the roaring 1990s, but not since. A more accurate measure of the information economy is certainly worth having, but it does not make America any less profligate.

Research cited:

This entry was posted in Economics. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *