 | | | | by Phil Mullan |
'You can see the computer age everywhere but in the productivity statistics.' This much-quoted claim by Nobel prize-winning economist Robert Solow was made in 1987 - reflecting the fact that, for a long time, it was impossible to find any sign of the IT age delivering above-average growth rates.
| Indeed, the supposed elevation of the role of information and the greater pace of technological change coincided with the post-1973 slowdown in economic growth. It is only more recently that the belief in a new information society has been given more explicit economic clothes.
| While the IT revolution got underway in the 1960s, with the spread of mainframe computing, any special impact on productivity growth could not be identified at least until the mid-1990s. 'Until that point, many studies had attempted to detect an increase in technical progress, but uniformly they met with no success', reports the company Goldman Sachs in Global Economics Weekly. Consequently, it explains, most economists reached the conclusion that 'the IT revolution was no more significant for productivity growth than previous bursts of technological progress - and indeed that it might be less important than the technological improvements which drove productivity growth to century highs between 1950 and 1972' (1). So for a long time, discussion of the economic consequences of the Information Age remained fairly muted. Every so often economists would address this so-called Productivity Paradox, but it had little wider purchase.
| Yet by the mid-1990s, the economic discussion in the USA started to become more upbeat. After two decades of productivity growth averaging one percent a year, US productivity jumped by nearly three percent in 1996. Economists started to talk of a new economic paradigm, and the IT revolution was believed at last to be delivering a New Economy.
| In early 1997 Bill Dudley and Ed McKelvey coined the title 'brave new business cycle' to describe a set of changes which they believed was taking place in the US business cycle. Rapid technological change was seen as a key factor underlying a smoother, less volatile, and possibly prolonged business cycle. Peter Schwartz and Peter Leyden popularised this perspective in an article called 'The Long Boom', printed appropriately in the magazine Wired in July 1997. 'We have entered a period of sustained growth that could eventually double the world's economy every dozen years, and bring increased prosperity for - quite literally - billions of people on the planet', they claimed. 'We are riding the early waves of a 25-year run of a greatly expanding economy that will do much to solve seemingly intractable problems, like poverty, and to ease tensions throughout the world. And we'll do it without blowing the lid off the environment.' (2)
| Not everybody was as breathless as Schwartz and Leyden in evoking IT as the new panacea, but the New Economy label has become widely accepted. The USA, at least, is regarded as having entered an historically significant wave of technologically enhanced growth; and this is universally assumed to derive from the rapid expansion of investment in IT equipment, more than tripling in the USA from 10 percent of business investment in 1960 to 35 percent in the late 1990s.
| | | A 'brave new business cycle'? |
| But despite all this hype, is there anything really exceptional in economic terms about the past five years? And even if something unusual is happening in the economy, and the five-year spurt will extend further into the future, how do we know this is primarily to do with IT?
| Business Week magazine might have described 'an epochal change in the history of production' (3), but real economic dynamism is still hard to find. The Japanese economy has been mired in recession for most of the past decade, the faltering common currency of continental Europe mirrors years of economic underachievement; and in Britain, productivity growth and investment have spent the past five years languishing below the long-term trend. And what about the 'resurgent' US economy? Even here, there is still scant evidence for the notion of an epochal shift upwards in productivity. Sure, the USA is today embracing its longest period of continuous economic growth, at about nine years, with no sharp recession in sight. Sure too, over the past five years, productivity has bounded ahead to an annual rate of almost three percent, after spending 20 years at an average of less than half that level. But when this recent spurt in productivity growth is compared historically, or when it is broken down into its components, the evidence for a nationwide revolution vanishes.
| For a start there is a statistical illusion to dispel. The comparison of a particular five-year period of growth with longer-term averages will usually tend to flatter the five-year period. This is because productivity growth moves in cycles. Growth in output tends to promote growth in productivity (output per hour) as spare capacity gets squeezed. The opposite happens when output declines, as employers tend not to get rid of labour at the first signs of economic slowdown. Productivity growth tends to follow and exaggerate the ups and downs of the business cycle. So it is not surprising that the second half of the 1990s - a time of steady though unexceptional economic growth - looks good in productivity terms in comparison with the longer post-1973 slowdown, which contains periods of both recession and expansion.
| More instructive is that for the discrete five-year period of 1995 to 2000, total productivity growth was little different to that of the period 1982 to 1987 - only a decade prior. And it is worth recalling that, during that period in the 1980s, some economists were already beginning to draw the superficial conclusion of a US renaissance based upon computerisation. At that time, though, this was greeted with scepticism from the likes of Robert Solow - and these economists' voices grew silent with the subsequent productivity slowdown at the end of the decade.
| So what is different today? In contrast to notions of an 'epochal change in the history of production', it seems that, since the 1980s, expectations have merely lowered. This conclusion is reinforced when we note that the level of the late 1990s productivity expansion, currently so acclaimed, is the same as the quarter-century average from 1950 to 1973. What was seen as normal for the post-war boom has, in a much shorter space of time, become apparent proof of a 'post-industrial' revolution. It is also worth remembering that US productivity growth during those 'Long Boom' years was itself nothing exceptional, falling below that of all its industrial competitors, including even the lacklustre Britain.
| Now let's turn to what makes up that 'impressive' 2.8 percent average annual productivity growth since 1995. Significantly, most of it comes from developments within the IT sector, not from the impact of IT upon the rest of the economy. Robert Gordon, an economist from North Western University specialising in productivity issues, attributes the bulk of the productivity growth since 1995 to higher productivity in the making of computers, rather than in their use. 'There has been no productivity growth in the 99 percent of the US economy located outside the sector which manufactures computer hardware, beyond that which can be explained by price re-measurement and by a normal pro-cyclical response', he claims (4).
| | | Real economic dynamism is still hard to find |
| In analysing the 'stunning' 5.1 percent productivity growth performance in the year to June 2000, Business Week, an unequivocal New Economy evangelist, had to come to a similar assessment (5). First, Business Week correctly noted that efficiencies in services, especially in financial operations, remain as difficult to measure as ever. In an economy two-thirds made up of services, this ambiguity in measuring tends to cast doubt on drawing any too-definitive conclusions from the aggregate statistics alone. All that can be said is that, so far as the official figures go, service sector productivity growth has remained well below the manufacturing rebound.
| In breaking down the easier-to-measure manufacturing sector, Business Week reported an even more revealing characteristic of contemporary economic activity. 'Data show that only three hi-tech industries - computers and office equipment, communications equipment, and semiconductors - account for all of the uptrend in factory productivity growth', stated the magazine. 'In the past year, productivity growth in these three sectors, which make up only 10 percent of total factory output, appears to have been in the order of 50 percent: output grew about 49 percent while hours worked declined about one percent. In the other 90 percent of manufacturing, the productivity miracle portrayed by the overall sector performance appears to be a lot less remarkable, despite the increased use of technology by many manufacturers.' The figure Business Week calculates for manufacturing productivity growth, excluding these hi-tech sectors during the second half of the 1990s, is about two percent: a little below its long-term trend level. So it appears that in the sectors where most IT investment is occurring - services and non-tech manufacturing - and where the technology-inspired business shake-up is supposed to be happening, there is no evidence even of a short-term productivity spurt.
| When breaking down the contributors to GDP growth in the USA, Merrill Lynch has similar findings. Over the past five years, on average, manufacturing has been growing at about five percent a year, but excluding the technology sector this falls to about two percent. So were it not for the New Economy effect, US growth would be pretty lacklustre. It is unlikely that this one dynamic sector, which relies primarily on the rest of the economy to buy its goods and services, can sustain such a growth rate gap for long.
| To summarise, the statistical picture in the USA is this:
| Business has been buying a lot of IT equipment, especially since the early 1980s, making up a much bigger chunk of total business investment. Stephen Roach, chief economist at Morgan Stanley Dean Witter, estimates that total spending on IT hardware, software, maintenance and management amounts to over 10 percent of US GNP. There is, therefore, no absolute shortage of resources committed to IT, and this has been the case for almost 20 years.
| Despite these high levels of investment, the statistics so far show minimal improvement in productivity in the sectors making the IT investments (in distinction to those making or supplying the IT) - especially in services where a disproportionate share of IT spending takes place.
| | | What was seen as normal for the post-war boom has become proof of a 'post-industrial' revolution |
| The reasonable overall economy productivity growth figures since 1995, showing an apparent return to pre-1973 long-term averages, do not reflect what is happening over all the economy. Rather they reflect rapid growth inside the IT sector, which is benefiting from this demand for its products. Recall that the cyclical pattern of output-productivity growth described above works at a sectoral level as well as for the whole economy. This is good news for those who own and who work in this sector, which makes up about five percent of the economy. But it is a far cry from illustrating an epochal economic transformation.
| Prophets of the New Economy doubt the accuracy of the official economic statistics, and especially what is happening in services - and it is valid for them to do so. But however else these statistics can be interpreted, they cannot be seen to show an economy-wide acceleration in productivity growth. You can claim that the figures are too stuck in 'old economy' assumptions - and this debate will rumble on - but you cannot legitimately deploy them as proof of a qualitatively superior level of economic performance, whether this is seen to be driven by new technology or by more traditional means.
| Even if one follows the more sensible of the 'it is too early to see' advocates of the New Economy, the future looks nothing like a third societal revolution. Another Goldman Sachs report forecasts an average annual increase in productivity of 0.4 percent above the level that it would have been without the information 'revolution' - amounting to, say, five percent over the next 10 years. Yet the impact of the agrarian and industrial revolutions, with which today's experience is being compared, saw increases in productivity not measurable in percentages but by orders of magnitude (6).
| This extra anticipated boost to growth is tiny even in comparison with the 'normal' level of growth capitalism has been able to deliver. During the twentieth century, with all the history of wars and the slump, national output in most industrialised economies grew by about 600 percent. For so many people to get excited over the prospect of an extra five percent on top of trend growth rates is puzzling. Even the economic impact of electrification - just one part of the ongoing dynamic of industrialisation in the twentieth century - was much greater, estimated at three times the forecast level for our New Economy and over a longer period, totalling a 25 percent boost to industrial productivity during the interwar years.
| But while we might not be seeing a qualitative uplift to economic growth, today's economy has changed in many ways, compared to most of the twentieth century. In that sense we could talk of a New Economy over the past 10 to 15 years - but one which has different features and determinants than are conventionally associated with the label. The experience of the genuine New Economy is characterised more by 'slow drift' than 'long boom'. In the 1970s and 80s, we had the return of a heightened form of the classical recession-recovery business cycle. But since the latter part of the 1980s, and especially since the 1990s, two related changes provided the backdrop for economic times that were more benign, though marked by a distinct lack of innovation or dynamism. In the industrialised countries, the disappearance of industrial conflict made possible a steady but relatively staid period of growth. During the same period, a more stable global political environment facilitated a more extensive international division of labour, and material production has tended to grow fastest in the newly industrialising economies. This has sustained a reasonable pace of world economic growth, though one marked by relatively languid growth in most of the older industrialised countries.
| The combined result of these domestic and international changes is that the business cycle, especially in the USA and Britain, has become less volatile and flatter. This, rather than anything to do with the specific phases of the development of IT, accounts for the recent disappointing record of productivity. IT itself is no more to blame for this genuine New Economy than it is for the fantasy one.
| Phil Mullan is the author of The Imaginary Time Bomb: Why an Ageing Population Is Not a Social Problem, IB Tauris, 2000 (buy this book from Amazon (UK) or Amazon (USA))
This is the second in his series demystifying the 'information society'.
Read on:
IT's potential: computing and communications by Phil Mullan
Don't Blow IT by Sandy Starr
spiked-issue: Don't blow IT
(1) Global Economics Weekly, 2 February 2000
(2) Peter Schwartz and Peter Leyden, 'The Long Boom', Wired, July 1997
(3) Business Week, 28 August 2000
(4) Robert Gordon, 'Has the "New Economy" rendered the productivity slowdown
obsolete?', Working Paper, Northwestern University, 14 June 1999
(5) Business Week, 28 August 2000
(6) See, for example, Peter Jay's recent The History of Wealth to get a sense of the genuinely epoch-changing dimensions of those two transitions
|
|
|
| | |
