Economists hold forth on ideas to boost productivity growth

May 19th, 2016 at 8:05 am

Paraphrasing Mark Twain’s famous comment about the weather, everybody complains about productivity growth but nobody does anything about it.

As OTEers are learning, for better or worse, I’ve been obsessed with the problem of slow productivity growth. It poses a fundamental constraint on the growth of living standards both here and in many other economies where the slowdown has been ongoing for a number of years now. Reading Robert Gordon’s magisterial new tome on the issue has only deepened my obsession, as Gordon is both highly authoritative on the issue and pessimistic about the path of future output per hour.* On the other hand, economic forecasting is particularly weak in this area and the experts have rarely, if ever, accurately predicted twists and turns in productivity growth.

One key insight I took from Gordon’s book is that productivity growth is more susceptible to policy than we think. The assumption that firms must be operating at the edge of their productive potential or they’d be out of business is surely wrong. There are inefficiencies being “left on the table.”

When unions in their heyday used their bargaining clout to raise their members’ pay, for example, higher labor costs meant firms either found ways to become more productive or shaved profit margins. So they found new efficiencies. (Josh Bivens and I have argued that this dynamic is likely operative in periods of full employment.) Gordon also vividly recounts the lasting impact of the war effort (WWII) in forcing productivity gains and lists numerous “headwinds” to productivity growth, including education, inequality, and poverty, all of which are variables that could be improved or worsened through policy or anti-policy (a.k.a. negligence).  He heavily stresses investment in human capital, starting with early childhood, as a way to boost productivity.

I’ve already offered a number of my own ideas on what might help, but this time I reached out to other economists to give me a sentence or more on what they think would be the best policies to raise productivity growth, either in the near or long term. Here are their (lightly edited) responses.

*[Be sure to listen to Gordon’s recent, excellent lecture on his book at the London School of Economics (here are the slides that go with it). Bob has his own interesting and elaborate policy agenda which I will write up separately. I could feed on this stuff all day but I don’t expect others to share that passion!]

Dean Baker, Co-director of the Center for Economic and Policy Research and co-author, with yours truly, of “Getting Back to Full Employment: A Better Bargain for Working People.”

I think full employment is the biggest deal. For me, that explains why productivity growth goes from ~2 percent pre-recession to less than 1 percent in the recovery. I don’t think there is going to be anything else that will buy you that much.

I do think we can get a lot by reducing/replacing patent and copyright protection (there’s incredible waste in these rents)” [JB: Gordon also stresses this last point re patents] “and also by downsizing the financial sector with a financial transactions tax. There are many other places where you could eliminate waste, most obviously health care, but none of them would produce anything like the bonanza from a full employment high wage economy.

One idea which may not help measured productivity much, but which I bet increases actual productivity hugely, is free public transit focused on buses (with designated bus lanes). If we got a large percentage of current drivers to instead take public transit it would enormously reduce the time wasted in traffic. That doesn’t show up in output per hour of work because we don’t count this time in productivity’s denominator [hours worked], but we certainly should in any reasonable measure of economic well-being.

Josh Bivens, Research and Policy Director, Economic Policy Institute

A big reason for the decline in last decade in productivity growth is reduced private capital investment. We should offset this with more public capital investment, including “core” infrastructure as well as renewable energy, education, and health care.

[Josh has also done great analysis of the linkages between full employment and productivity growth: “we need to seriously consider the possibility that productivity growth (normally thought of by economists as a supply-side phenomenon) is just the last casualty of the chronic demand shortfall that brought on the Great Recession and which was never filled in sufficiently to push the economy back to full health.”]

Alan Blinder, Gordon S. Rentschler Memorial Professor of Economics and Public Affairs at Princeton University

I’d go for early childhood education, especially for less advantaged kids. It takes a long time, but the payoffs are impressive.

[He’s surely right about that, and we do a lot less of this than other advanced economies.]

Heather Boushey, Executive Director, Washington Center for Equitable Growth and author of the new book, “Finding Time: The Economics of Work-Life Conflict” (Harvard University Press).

If we’re concerned about productivity, we have to be concerned about both the rate of technological progress and the quality and quantity of labor supply. In the United States, employment rates for both men and women have fallen both absolutely and relative to our economic competitors in the OECD. The United States now ranks 19 out of 21 OECD countries in primary age female labor force participation and there is evidence that this is in no small part because of our lack of attention to policies that allow families to address conflicts between work and life. The United States stands alone among the OECD for not providing paid parental leave, for example, although four states—California, New Jersey, Rhode Island, and New York—now have implemented programs. Schedules are also an issue and thus it’s important that this Wednesday, the U.S. Department of Labor announced updated overtime standards, which also are the kind of policy that can reduce the day-in, day-out conflicts over time use and help boost or maintain labor supply.

[Bonus: JB on new overtime rule.]

John Fernald, Senior Research Adviser at the Federal Reserve Bank of San Francisco

[John is one of today’s foremost experts on growth theory and evidence, and I noted his measured expectations re the impact of policy on productivity growth (“no magic bullets,” “at the margins,” “slightly”). It’s an important reminder that a lot of what drives the productivity trend remains elusive.]

There are no magic bullets that would restore rapid productivity growth. But, at the margins, sensible policies around infrastructure, education, basic research, and balanced regulation can help. Many policies in these areas complement private-sector investments in capital and innovation; they pass microeconomic cost-benefit tests even in a low-growth world.  Because of this complementarity, these policies (slightly) raise the probability of seeing a larger, broad-based break-through in growth.

Jason Furman, Chair of President Obama’s Council of Economic Advisers.

The biggest source of the recent productivity slowdown is the slowdown in growth of private investment (which, by the way, has happened across the advanced economies), which has resulted from the overall slowdown in global demand–when demand is weak there is less reason for investment. We all like more investment in public infrastructure, but an underappreciated benefit is that by strengthening aggregate demand it would also increase the incentive for private investment as well, boosting productivity growth. When demand is low, expanding public investment can “crowd in” private investment. In the medium- and long-run, however, variations in productivity growth are almost entirely driven by “total factor productivity” (or innovation), not by added capital. Expanding the tax credit for research and development is one way that business tax reform could help productivity growth, essentially helping businesses to take into account the positive spillovers (“positive externalities”) that their investments in research have on the economy more broadly.

[Jason’s third link above takes you to the infrastructure chapter in his squad’s latest Economic Report to the President. For those interested in this topic–and if you’re still with me, that’s you–it’s an important read–a great survey of the lay of the land and the empirical connections between such investments and growth.]

Alan Krueger, former chair of President Obama’s CEA and Bendheim Professor of Economics and Public Affairs at Princeton

In addition to the usual suspects of investing more in R&D and corporate tax reform, I’d recommend considering lowering the threshold for overtime pay from 40 hours a week to 35 hours a week.

There are a couple of reasons why I think that lowering the OT threshold will raise productivity. First, companies will use workers’ time more efficiently if they have to pay more for the marginal hour or if they have a shorter workweek. Second, workers are fatigued with long hours. 

It is also worth noting that work hours have declined ‎more strongly in other countries as income has risen than in the US, so, in some sense, I think we have been facing a market failure when it comes to work hours. Our labor market creates a rat race where workers put in more time at work than is socially optimal, and the last hours are not especially productive.

[To be clear, the “OT threshold” to which Alan refers is not the salary threshold that was just raised by the Obama administration. It’s the 40 hours/wk threshold above which covered workers must be paid time-and-a-half.]

Erica Groshen, Commissioner, Bureau of Labor Statistics

Improving access to government administrative data so that statistical agencies could have access to more data to produce more and better official statistics would improve productivity in three ways: reducing burdens on respondents, improving the values of government statistics, and (likely the largest, but hardest to measure) improving the business, policy and personal decisions (allocation, investment, etc,) in the economy.

[It’s also important to be able to pursue] what we call “data synchronization”–allowing BLS and Census to share our establishment lists so that we can assign industry codes consistently across agencies. [Doing so could improve the accuracy of the] GDP, productivity and most national accounts.

[To learn more about the Obama administration’s statistical agencies’ ideas in this space, see this recent chapter on the role and use of “administrative data.” Note that these are data we are already collecting but not using – again, leaving efficiency gains on the table:

“Administrative data are data collected by government entities for program administration, regulatory, or law enforcement purposes. Federal and state administrative data include rich information on labor market outcomes, health care, criminal justice, housing, and other important topics, but they are often greatly underutilized in evaluating programs’ effects, as well as in day-to-day performance measurement and for informing the public about how society and the economy are faring.”]

Mark Thoma, Professor of Economics, University of Oregon

My first idea is more vigilant antitrust enforcement. As Dietrich Vollrath recently argued, a firm with a powerful market position has less incentive to invest in innovation. In addition, firms with market power can restrict innovation by other new or existing firms through strategies that limit their ability to enter markets.

The second is to do more to ensure that disadvantaged groups such as women and minorities can reach their productive potential, or have equal access to the resources needed to become entrepreneurs when they have innovative ideas. For example, to a large degree venture capital does not appear to be as readily available to women and minorities as it is to men.

Third, historically government investment in basic research has led to many productivity enhancing offshoots. Presently, support for basic research is at a 50 year low. A reinvestment in basic research could pay big dividends.

Finally, recent research has documented a falloff in the startup of ‘transformational entrepreneurial firms’ that have made substantial contributions to economic growth in the past. The reasons for this aren’t known, but finding the answer, and then implementing policies that can help to reverse this trend, could help to overcome growth stagnation.

[Re this last point, one wonders if there’s a size-driven acquisition story embedded in here somewhere, related to Thoma’s first point about antitrust enforcement: maybe large, rich firms absorb the small ones before they can flourish (e.g., Skype getting swallowed by Microsoft).]


As Fernald notes, no silver bullets but lots of common threads re public investments in both physical and human capital, along with less commonly heard ideas about work hours, work-family policies, and better data.  Nobody mentioned functional government, but I’m going to stipulate that this is implicit in all entries.

More to come!

Print Friendly, PDF & Email

9 comments in reply to "Economists hold forth on ideas to boost productivity growth"

  1. A concerned citizen says:

    My personal productivity decline? I have to manage healthcare, retirement, and a variety of other things, instead of being productive at work. All of these responsibilities have shifted to me over the past 5 years.

    Instead of a co-pay, now I have a HSA. I need to “shop” for decent providers and prices. Shopping takes time. After insurance processes claim, I have to make sure that they did it correctly, then pay the residual amount. For a family of 4.

    For rental property, just getting someone to answer the phone for handyman jobs takes a lot of time.

    At my job, similar shifts in responsibility. As a Senior Engineer, I have to use particularly slow web-based software to perform just about every function. Scheduling, admin, and reporting duties are now my responsibility over the past 5 years.

    From the high level economics perspective, there is no extra $ in the middle class to spend. I’m saving every $ for my daughters college education. But to be fair, I’ve never been a big consumer; I prefer outdoor activities rather than new cars or luxurious vacations.

    Not a big mystery to me.

  2. Amateur says:

    Thanks for a great read.

    I’m a bit quirky in that my biggest concern is that we’re measuring this properly. I’m glad you brought in the BLS so we could hear their view.

    I’m not sure how an economist can claim that we are measuring labor productivity properly when standard accounting doesn’t count foreign labor as labor. It doesn’t. How many times do I have to repeat myself? Can’t somebody respond? Am I just so obviously wrong that you are all embarrassed for me or don’t think this deserves a response? Or am I right?

    • Smith says:

      Good point, which John below also compliments you on, and to clarify the point further about foreign labor, America is in a productivity death spiral in certain industries. If American companies invest in new (more productive) factories in other countries, how does productivity rise. Yes they may close the least productive factories here, but they may also be the only productive factories here. Walmart may have already reached the saturation of productivity, whereas factory production has not, but it occurs some place else. Contrast that to the past when job growth wasn’t in low wage service sector.

      To the threatened rust belt worker, auto worker, assembly line worker, it doesn’t matter who pays for a wall, or whether the price of cars, textiles, and electronics go up, job security trumps other considerations.

  3. Roger Chittum says:

    Well this is off the wall, and I’m not going to try to prove it, but will assert for the sake of argument that one way to improve productivity is to reduce the drag on it caused by devoting evermore resources to digital data security. It seems to me that the highest rate of increase in productivity from digitization and networking is years behind us and that now all the new technical labor is going into protecting (and hacking) the data. Productivity growth in the digital world may currently even be negative as a result. I have no solution to propose. I just suggest it is a baked-in drag on productivity that must be overcome.

  4. Shaun says:

    I’m curious, could the widespread adoption of more tech actually be productivity destroying, rather than enhancing? For example, the proliferation of social media, the internet, and various other communication apps have become giant time sucks and probably distract from work more than enhance it. I really don’t have any data to back this up, but it seems we’ve gotten rather distracted as a society. Again, I only have anecdotal evidence and my hunches, but I’d be interested to hear your thoughts.

  5. Tom Cantlon says:

    Not to be contrary but just to keep perspective, GROWTH of productivity is interesting, and if it’s benefits get spread equitably then all benefit, but just for a hypothetical, what if productivity just froze? What seems to be more critical to the general state of most people in the economy is the distribution. That gets back to your other favorite topic, full employment, which I know you’re saying is linked to productivity too. But what’s the benefit of increased productivity if we don’t make serious headway at distribution problems? In fact, increased productivity might exasperate those problems, at least in the short run. The more productive we are, the fewer total work hours are needed, unless we intend to consume the planet in an effort to keep everyone at full employment. The more we achieve that state of not needing everyone working full time to produce what we want, the more we need a market arrangement that pays people a living for fewer hours of work. One that distributes the benefits of good productivity. We can’t even distribute the benefits we have now. Greater productivity is only going to add to that imbalance. It would seem the much greater concern is distribution. Or if I’m not seeing the reason why productivity should get such attention, then maybe the reason for it could be spelled out better? I know, productivity growth means economic growth, but that just brings me back to the start of this comment.

  6. john says:

    There is a good story why the productivity series is not comparable from beginning to end.

    Basically, we used to apply capital to labor here in the USA. But we have largely shifted the marginal application of capital to labor located outside the USA.

    Yet, the start of this process, at the maquiladora plants just over the border in Mexico, was fully and clearly owned by US investors, explicitly using law to arbitrage US labor costs. What has happened since then has been of exactly the same character except that the form of ownership has changed. Plants were built in China that were nominally locally owned, but their entire revenue came from a single US customer. That’s fictional independence, an accounting nicety. By now, US investors have sufficiently developed foreign owned outsource firms that it looks like we are just hiring them for a job. But we created them.

    To make the series comparable over time all this foreign labor and capital content would have to be artificially dragged back into the US numbers. Good luck with that exercise.

    I’m with Josh, it’s the application of capital to US labor that determines US productivity. And “Amateur” above.

  7. Smith says:

    My take on factors influencing productivity is much different than the opinions expressed above. I don’t think the issues raised are at the margins, and stuff like R&D are silver and magic bullets.

    1) Slow productivity growth is really only a post recession phenomena, because 1995-2005 it grew at 2.98% average.
    2) Implied and spoken of above, but not stated as clearly, does higher wages give incentive for productivity gains or as many argue, higher wages must only follow productivity gains. I’d guess a bit of both with emphasis on the former, an extremely significant distinction.
    3) Surely the drop off in R & D, both private and public would lead to an expected decrease in productivity gains. Likewise lack of antitrust has to stifle innovation.
    4) A wholesale attack on patents and copyrights is misguided and dangerous when previous protection served the country so well. What needs to be addressed are changes dictated by corporate greed, the ever lengthening of copyrights, the abuse of patents by big pharma, the use of government research money and discoveries which become privatized, the change from “first to invent” to “first to file” which rips away previous protection from independent inventors, as well as other recent changes to benefit corporations.
    5) Of course the lack of work life balance and most modern office workers still exempt status has a huge effect on productivity. Think about it, the unrecorded hours should increase the productivity number. It doesn’t, and the 35 hour work week should be advanced for everyone.

    (and as if on cue : )

  8. Vintage Rocker says:

    Tough call. I agree with the author and some commenters that the real deal is innovation and technological progress, and Bob Gordon is the ultimate source on how it has stalled in the last decades, albeit not necessarily why.

    My own take: , although diagnosing the causes is much easier than proposing a cure. Less regulation seems attractive, until you consider the potential costs. a UBI to release time for more creative, risky endeavors is probably the closest we have collectively to a silver bullet, but even then it’s a risky proposition