Thursday, November 27, 2014

The Unstarvable Beast

CAMBRIDGE – As the world watches the United States grapple with its fiscal future, the contours of the battle reflect larger social and philosophical divisions that are likely to play out in various guises around the world in the coming decades. There has been much discussion of how to cut government spending, but too little attention has been devoted to how to make government spending more effective. And yet, without more creative approaches to providing government services, their cost will continue to rise inexorably over time.

Any service-intensive industry faces the same challenges. Back in the 1960’s, the economists William Baumol and William Bowen wrote about the “cost disease” that plagues these industries. The example they famously used was that of a Mozart string quartet, which requires the same number of musicians and instruments in modern times as it did in the nineteenth century. Similarly, it takes about the same amount of time for a teacher to grade a paper as it did 100 years ago. Good plumbers cost a small fortune, because here, too, the technology has evolved very slowly.

Why does slow productivity growth translate into high costs? The problem is that service industries ultimately have to compete for workers in the same national labor pool as sectors with fast productivity growth, such as finance, manufacturing, and information technology. Even though the pools of workers may be somewhat segmented, there is enough overlap that it forces service-intensive industries to pay higher wages, at least in the long run.

The government, of course, is the consummate service-intensive sector. Government employees include teachers, policemen, trash collectors, and military personnel.

Modern schools look a lot more like those of 50 years ago than do modern manufacturing plants. And, while military innovation has been spectacular, it is still very labor-intensive. If people want the same level of government services relative to other things that they consume, government spending will take up a larger and larger share of national output over time.

Indeed, not only has government spending been rising as a share of income; so, too, has spending across many service sectors. Today, the service sector, including the government, accounts for more than 70% of national income in most advanced economies.

Agriculture, which in the 1800’s accounted for more than half of national income, has shrunk to just a few percent. Manufacturing employment, which accounted for perhaps a third of jobs or more before World War II, has shrunk dramatically. In the US, for example, the manufacturing sector employs less than 10% of all workers. So, even as economic conservatives demand spending cuts, there are strong forces pushing in the other direction.

Admittedly, the problem is worse in the government sector, where productivity growth is much slower even than in other service industries. Whereas this might reflect the particular mix of services that governments are asked to provide, that can hardly be the whole story.

Surely, part of the problem is that governments use employment not just to provide services, but also to make implicit transfers. Moreover, government agencies operate in many areas in which they face little competition – and thus little pressure to innovate.

Why not bring greater private-sector involvement, or at least competition, into government? Education, where the power of modern disruptive technologies has barely been felt, would be a good place to start. Sophisticated computer programs are becoming quite good at grading middle-school essays, if not quite up to the standards of top teachers.

Infrastructure is another obvious place to expand private-sector involvement. Once upon a time, for example, it was widely believed that drivers on privately operated roads would constantly be waiting to pay tolls. Modern transponders and automatic payment systems, however, have made that a non-issue.

But one should not presume that a shift to greater private-sector provision of services is a panacea. There would still be a need for regulation, especially where monopoly or near-monopoly is involved. And there would still be a need to decide how to balance efficiency and equity in the provision of services. Education is clearly an area in which any country has a strong national interest in providing a level playing field.

As US President in the 1980’s, the conservative icon Ronald Reagan described his approach to fiscal policy as “starve the beast”: cutting taxes will eventually force people to accept less government spending. In many ways, his approach was a great success. But government spending has continued to grow, because voters still want the services that government provides. Today, it is clear that reining in government also means finding ways to shape incentives so that innovation in government keeps pace with innovation in other service sectors.

Without more ideas about how to innovate in the provision of government services, battles such as one sees playing out in the US today can only become worse, as voters are increasingly asked to pay more for less. Politicians can and will promise to do a better job, but they cannot succeed unless we identify ways to boost government services’ efficiency and productivity.

  • Contact us to secure rights


  • Hide Comments Hide Comments Read Comments (11)

    Please login or register to post a comment

    1. CommentedOliver Kovacs

      Reading Prof. Rogoff's articles is always instructive.
      As the empiria showed, neither the fiscal stimuli nor the fiscal austerity could trigger growth impact, instead, fiscal conditions have just worsened further due to the decline in GDP. A some type of aurea mediocritas consolidation should be of paramount importance that addresses the issue of supporting R&D&I in a pro-cyclical way while counter-cyclically reduces expenditures in inproductive spheres (public sector wages, salaries, social transfers as Alesina and other works on non-keynesian effects justified).
      I would raise the issue of how to incentivise public sector to be more innovative in favouring the term "more for less". Innovation, as it was rightly pointed out by many, is hampered by a lot well-documented factor, however, the literature does not devote enough attention to the importance of inherent incentives of innovation that differ across institutional architectures. /see a policy brief on "Policies Supporting Innovation in Public Sector Provision" which tries to address this crucial gap /

    2. Commentedjack lasersohn

      While it is certainly true that 'voters want the services that government provides' it is equally true that most of them do not bear the cost of those services, which is shifted to a tiny fraction of the population through progressive taxation. As a result, and exactly as in the market for healthcare, the demand for essentially 'free' service increases without limit.
      Moreover, most government expenditures are transfer payments, which arguably have experienced nearly exponential productivity growth over the past 50 years as it takes virtually the same number of labor hours to process checks for $1 trillion as for $1 million.
      Also, as you point out, the growth in 'productivity' in Defense has also increased dramatically.
      The areas of government where productivity has remained stagnant, like education, are still relatively small and not really relevant to the problem of government growth.
      The real problem is in the growth of entitlements, where certain voters have learned that they can force others to pay for services they desire.
      That is the core problem of all democratic systems and has nothing to do with lack of productivity growth in government (although of course it would be less of a problem if we had faster productivity growth overall).
      There is no obvious solution to this problem, except that it will stop when it reaches some natural limit , as in parts of Europe.

    3. CommentedCharles Broming

      I agree with Prof. Rogoff's analysis at the highest level, viz., the service sector's problems with productivity, cost and, therefore, price. I agree as well that we need to change our conversations from exhortations to reduce government spending to conversations about how to deploy government funds more effectively. But, the issue needs to be framed appropriately and the news media needs to observe and report on it. There has been plenty discussion about government spending effectiveness over the decades, but the news media have ignored it. It's complicated, contentious, doesn't offer good sound bites and is, therefore, hard to cover and offers lower returns.

      The fundamental questions that need to be addressed (and probably never answered completely or finally) are, "What is the "right" size of government (at which level)?" and, "Which services do governments provide more effectively and efficiently than the private sector over the long run?"

      The second question can be answered; the tools are available. A credible and reasonable answer to the second question would be a product of an answer to the first question. To argue for reduced spending based on some prior faith in the priority of "limited government" puts the cart before the horse. In point of fact, those who argue for "limited government" have no opponents in America or Western Europe, and I doubt that even the most totalitarian dictators (kings, princes, etc.) believe that "unlimited government" is a possibility, much less an alternative.

    4. CommentedMichael Scheps

      Surprised that Professor Rogoff would be so wrong in his evaluation of Reagan's "starve the beast" philosophy. David Stockman, after leaving the OMB, wrote in his book "The Triumph of Politics: Why the Reagan Revolution Failed", that they failed to understand Congress would not curtail spending in the face of reduced revenue. That, and the Reagan policy of outspending the Soviet's on defense, were the 2 primary reasons that Reagan left office with a large national debt that was not addressed until President Clinton's administration.

        CommentedMichael Scheps

        Mr. Bromberg,

        Thank you for verifying my point. Professor Rogoff wrote- "In many ways, his approach was a great success". I believe that a historical evaluation proves exactly what you stated- "pragmatic aspects of politics, especially re-election and post-congressional employment opportunities, were more important to members of Congress than was Reagan's ideological crusade". Reagan's approach might have been admirable, if that is ones political philosophy, but in reality it didn't, and up to now, hasn't worked.

        CommentedCharles Broming

        Mr. Scheps,

        Prof. Rogoff's account of Reagan's philosophy is accurate. Stockman (in his book and his Fortune magazine article) simply pointed out that the pragmatic aspects of politics, especially re-election and post-congressional employment opportunities, were more important to members of Congress than was Reagan's ideological crusade. Thus, the real, "don't-tax-and-spend" Republicans emerged.

    5. CommentedProcyon Mukherjee

      Central Bank actions had become virtually fiscal in nature and now we have seen that monetary release did not find its way in goods and services, if at all it had it had increased stocks of unused houses, or inventory and piles of commodities from Aluminum to many other forms.

      Uncertainty channelized investments into unproductive ‘investment assets’, which are a parking lot for ‘certainty’ to return, the attractiveness of such assets like commodities, stocks or bonds runs against the unattractiveness of real economy ‘options’ that produce goods and services that get consumed to create jobs. This is now becoming a permanent feature. So we have already a stock waiting to be consumed and then we are venturing into further fiscal stances that would make a push for further debt escalation as there is no shortage of funds but shortage of viable channelizing options into goods and services that create jobs.

      Is the government a better bet for this arrangement to continue in form of government spending that is financed by cheap debt again? Let us take stock of what has already been spent in the last three years and how many net jobs it actually created on a permanent basis.

      Procyon Mukherjee

    6. Portrait of Pingfan Hong

      CommentedPingfan Hong

      "If people want the same level of government services relative to other things that they consume, government spending will take up a larger and larger share of national output over time.": this is not true. Because people want an increasing level of public services relative to other things that they consume, government spending increases its share in the economy over time. Public healthcare is a good example.

    7. CommentedTim Chambers

      Privatization of government services is not the solution. What you advocate is rent extraction from services that are done well enough by government employees. There are far too many examples of companies handling this kind of work that waste enormous sums to raise profits on cost plus ten percent contracts.

      As for improving teacher productivity, it isn't going to happen. Computer programs might be good at finding spelling and grammar errors, but they can't tell a student how to do a better job of researching or elucidating his topic. That is something only a human can do.

      The real problem that we have is that by de-emphasizing manufacturing, we have off-shored the growth of wealth creation that supports the service sector, which simply moves existing wealth around. That wealth, in all too many cases, is not being returned to this country because the companies that create it refuse to pay taxes on it.

      Too many of the "wealth creators" demand a free lunch, and far too many of their current and former employees are being subsidized by the government because their economic circumstances have been so drastically reduced via free-trade agreements with third world countries.

      Neo-liberalism is the problem, not the solution.

    8. CommentedVictor Stern

      I disagree with the diagnosis. Education is not breaking the fisc. Its medical services, where increases in productivity increase demand to no natural boundary. We would all chose to live forever together, if we could.

      Also disagree with application of the "cost desease" concept. A significant portion of growth in the share of services in the economy may be due simply to monetization of services previously rendered within households and communities without exchange of money.