Today, about four out of every five dollars spent in the OECD economies purchases services or intangible goods. This “dematerialization” of economies demands a more nuanced understanding of what drives productivity.
The word “productivity” typically calls to mind industrial assembly lines pumping out cars or washing machines, breakfast cereal or shoes. The word may also conjure images of crops being harvested, livestock being butchered, or houses being built. It is less likely to elicit thoughts of haircuts, streaming television, or mortgages. Yet nowadays, it is largely these kinds of intangible goods and services that define economies.
Many economists equate “total factor productivity” with technological progress. Northwestern University’s Robert Gordon, for example, predicts that productivity growth will continue to slow – as it has done in most developed economies since the mid-2000s – because today’s digital innovations are, in his view, less transformative than earlier advances like the flush toilet, radio, and the internal combustion engine.
But, today, about four out of every five dollars spent in the leading OECD economies purchase services or intangible goods. This “dematerialization” of economies – which I observed in the 1990s, and which figures like digital economy expert Andrew McAfee have lately been exploring – is complicating our understanding of productivity.
In fact, in much of today’s global economy, even the production of tangible goods is shaped by a growing number of intangible factors. As Seth Lloyd of the Santa Fe Institute has pointed out, a farmer hedging against bad weather or disease now operates largely in the realm of ideas.
Whereas in the past, farmers would “insure” against the failure of one type of crop by planting others or raising livestock – that is, through physical diversification – today they do so largely by applying agricultural science, like testing soil and assessing climate conditions, or even by participating in options markets. Such intangibles – in addition to new technologies, such as irrigation – produce the discrepancies McAfee observes in crop tonnage produced from the same amounts of inputs.
Still, when it comes to agriculture, the end result is easily quantifiable. That is not the case for many other modern productivity-boosting innovations.
In a recent presentation, Leonard Nakamura of the Federal Reserve Bank of Philadelphia offered several examples, including energy-efficient buildings, lane-keep-assist and parking sensors in automobiles, and GPS navigation. Innovations in health-care treatment also qualify. For example, using the cancer drug Avastin to treat macular degeneration is far less expensive than using Lucentis, one of the drugs originally approved for that purpose.
In theory, the effects of some of these innovations on productivity could be quantified through quality-adjusted pricing. Cars with sensors that facilitate parking and improve road safety might be discounted, resulting in a higher “real” measured output for cars.
But, in practice, such adjustments pose a significant statistical challenge, owing to the pervasiveness of the underlying technologies. Consider GPS navigation: how do you quality-adjust for the use of apps like Waze or Google Maps?
When it comes to medical, legal, and other professional services, quantifying productivity is even trickier. One approach focuses on outcomes – say, a longer career (thanks to better health care) or higher profits (thanks to management consultants).
But these improvements cannot be traced back to a single factor. Doctors and hospitals are essential to extend people’s healthy lives, but so are living conditions, diet and exercise, social connections, and even having a pet. Luck – for example, not being exposed to a disease outbreak – also plays a role.
Some of my University of Cambridge colleagues are working to deepen our understanding of these dynamics by examining the connections between social capital and productivity. This approach – which reflects a shift toward a broader view of productivity – is a step in the right direction.
This conclusion seems to be borne out by history. As Corinna Schlombs of the Rochester Institute of Technology shows in her new book Productivity Machines, in the twentieth century, one of the key differences between the approach of American industrialists and productivity experts and that of their European counterparts was that the latter were more likely to view productivity in purely technical terms.
After World War II, during the Marshall Plan era, Americans showed visiting European workers and industrialists new ways to organize production. (The assembly line is as much an idea as a technology.) Moreover, they touted America’s more egalitarian social dynamics, including its public school system and broad civic involvement. The recognition that “soft” innovations were at least as important as “hard” technologies, Schlombs suggests, was the decisive factor behind America’s superior productivity.
So perhaps today’s pervasive productivity slowdown should not be blamed solely on an unsupportive macroeconomic environment, let alone on inadequate technological innovation. (Software engineers and biomedical researchers would scoff at the latter notion.) Social and cultural contexts that are fragmented, unequal, or otherwise problematic may also be playing a role.
Project Syndicate
