Last December I posted about the recent slowdown in US productivity growth, and whether the party of IT-fueled growth really was over. I hypothesized that it wasn’t, but concluded that:
"Of course, if the data continue to show productivity declines I’ll have to revisit my optimistic hypotheses and opinions…. Let’s all stay tuned — I can’t think of a more important number for us to watch and discuss."
Well, the most recent data have not been kind to us optimists. Annual nonfarm productivity growth (seasonally adjusted) in the first quarter of 2007, was only 1.0%, according to recently revised estimates, which is less than half the corresponding figure from Q4 2006. This is quite far below the annual average growth rate of 3.87% the US economy experienced from Q3 2001 to Q2 2004.
So even though many expect productivity growth to pick up somewhat in the second half of the year, the trend is clearly down from its peak.
Does this indicate that the era in which information technology was having a deep impact on the US economy has passed? Evidence in support of this view comes not only from the productivity statistics, but also from the slowdown in IT investment growth rates themselves. Taken together, they tell a straightforward and intuitively appealing story: companies are making rational decisions to stop spending willy-nilly on IT because they realize that IT is doing less for them now than was the case in the past.
A logical conclusion of this story is that absent any radical tech innovations, this gradual cooling will continue. We’ll continue to spend less and less on IT if IT does less and less for us.
For the reasons I outlined in my December post, though, I’m still optimistic that many, many opportunities remain for IT to increase corporate productivity. Last week, for example, I visited Rearden Commerce, where I saw a demo that made my mouth water. Rearden wants to take the hassles, busywork, and inefficiencies out of processes like making travel and dining reservations, setting up Webex sessions and conference calls, and shipping packages. These are far from core processes, but Rearden’s demo showed me how much opportunity there is to do them better and faster, and so to put hours back in the weeks of many of us knowledge workers. We’ll use at least some of those hours to generate more output, thereby increasing productivity.
Rather than continuing to scrutinize the productivity figures, I’d like to shift topics. Everyone agrees that productivity growth is a critical measure, but it’s not the only one managers care about, and it’s not the only one that could be influenced by IT. Productivity growth measures whether the economic pie is getting bigger, but it’s also important to understand how that pie is being divided up.
Productivity growth, in other words, doesn’t tell us anything about competitive balances or competitive dynamics. And it’s perfectly possible for IT to have no impact on aggregate productivity at the same time that it’s having a substantial impact on competition. To see how, consider a stripped-down industry with only two firms, both of which have 50% market share. One of them invests successfully in IT and uses its new technology-based capabilities to take customers from the other company. After a couple years one firm has 75% market share, the other 25%, but the industry remains the same size during this period because total customer demand doesn’t increase. At the industry level, IT has had no impact on productivity in this example (since total output doesn’t increase) but it’s had a major impact on relative competitive position.
In the real world, of course, IT typically impacts both productivity and competition. The point of the example is simply to show that the two don’t have to move in lockstep. So far, large-scale research has concentrated on IT’s link to productivity. This scholarship has been very valuable, giving decision makers confidence that their IT investments are boosting output.
My colleagues and I think it’s now time to apply similar research methods to investigating IT’s links with competition. MIT’s Erik Brynjolfsson, HBS’s Michael Sorell and Feng Zhu, and I wrote a paper examining the association between IT investment and two industry-level measures of competition: concentration growth (is the industry increasingly subject to winner-take-all dynamics?) and turbulence (do companies maintain the same rank order year after year, or do rankings jump around a lot?). We found strong and positive relationship between IT and both of these outcomes. Erik and I discussed these results in a recent article that appeared in the Wall Street Journal and Sloan Management Review.
More recently the four of us have been examining whether companies in IT-intensive industries tend to cluster tightly around a single average level of performance (an ROA of x%, say, or an EBITDA margin of y%, or z dollars of market capitalization for every dollar of revenue), or whether there’s a wide variation around this average. Our results are very clear, and I find them fascinating. I’ll discuss them later; for now, let me just say that they have convinced me more firmly than ever that IT matters.
So even if the IT-fueled productivity boom is over, IT-fueled competition isn’t necessarily winding down. Our results, in fact, indicate that it’s not. When I wrote in December that I couldn’t think of a more important number to watch than the quarterly productivity growth figure, I wasn’t thinking hard enough. We should also be keeping a weather eye on measures of competition, and the strength of the links between them and IT.
One final point: IT-fueled competition could continue long after IT investment growth rates slow down. All the ‘building blocks’ of an IT infrastructure are continuing to get cheaper over time, so it would be remarkable if total IT spending didn’t slow down at some point. It would be very dangerous, however, to assume that when this slowdown occurs it means that corporate strategists can go back to their ‘regular’ jobs and stop thinking about how IT can help them get ahead of their rivals.