Growthology

Growing Pains

My inaugural column for Forbes.com, "Avoiding the Coming Growth Slowdown," is now up.  Here's a teaser:

Consider this puzzle. Compare economic performance in two periods: 1973-1990 versus 1990-2007. Both periods are 17 years in length; both begin and end with the last year of an economic expansion. In the earlier period, the U.S. economy weathered oil shocks, stagflation, and a punishing recession in the early ’80s, and growth in labor productivity in the nonfarm business sector limped along at a dismal 1.33 percent a year. In the latter period, prosperity was interrupted only by a pair of brief, mild recessions, and the IT revolution led a dazzling rebound in productivity growth up to 2.33 percent a year. Yet in 1973-1990, real gross domestic product per capita rose at an average annual rate of 1.93 percent – better than the 1.85 percent average annual growth rate during 1990-2007. How could that have happened?

Check out the whole thing.

Raising Productivity Growth

Courtesy of the McKinsey Global Institute, here is an interesting new contribution to the debate kicked off by Tyler Cowen's The Great Stagnation. This new MGI report -- entitled "Growth and renewal in the United States: Retooling America's economic engine" -- doesn't mention Tyler's book, but it offers evidence that supports his argument -- as well as that of his critics. On the one hand, MGI makes a strong case that we are facing strong headwinds when it comes to prospects for future growth. On the other hand, contrary to Tyler, the report points out that there's plenty we can do about it.

The MGI report singles out demographics as the main obstacle to future growth, and it makes a strong case. Think about per-capita GDP growth as a function of two variables: growth in output per worker-hour (i.e., productivity growth) and growth in total worker-hours per capita. If the latter slows down, the former has to speed up for growth to stay on trend. According to MGI:

Today, however, the contribution from labor is slowing down as baby boomers retire and the female participation rate has plateaued. In the next ten years, the proportion of working-age Americans will decline from 67 percent to 64 percent. By the 2020s, the contribution of labor to US GDP growth rates is expected to decline to just 0.5 percent from a peak of 2.0 percent in the 1970s.

Accordingly, to avoid a dropoff in the long-term growth rate, labor productivity growth needs to compensate for the unfavorable demographic situation. Specifically, the MGI report says that output per worker-hour would need to increase by nearly a quarter to maintain the historical average of 1.7 percent annual growth in real GDP per capita.

Unlike Tyler, though, the folks at MGI don't think the orchard of "low-hanging fruit" is bare.  They see plenty of opportunities to boost productivity: diffusion of existing best practices, implementation of the next wave of emerging business innovations, and restructuring of low-productivity and highly regulated sectors (namely, education and health care) all offer promising gains. 

In addition, the report helpfully points out that our demographic situation is not unalterable. If the United States could achieve the same labor force participation rates for women and workers 55-64 as Sweden, the same youth unemployment rate as the Netherlands, and the same immigration rate as Australia, it could expect to add a full percentage point to its real GDP growth rate over the next 10 years. 

On a critical note, the MGI report is sometimes frustratingly short on specifics. In particular, it's much too vague in addressing what needs to be done to boost the abysmal productivity record of the health and education sectors. Nonetheless, it does frame the issue well -- and points the discussion in a more constructive direction than Tyler's fatalism.

The Great Stagnation?

That's the title (sans question mark) of a new e-book by Tyler Cowen, economics professor at George Mason University and blogger extraordinaire at the wonderful Marginal Revolution. The thesis is summarized by the book's fashionably prolix subtitle: "How America Ate All the Low-Hanging Fruit of Modern History, Got Sick, and Will (Eventually) Feel Better."

Tyler argues that, ever since colonial times, the American economy has benefited from "low-hanging fruit" -- i.e., bountiful opportunities for growth. He singles out three in particular: free land, technological breakthroughs, and smart but uneducated kids. "Yet during the last forty years, that low-hanging fruit started disappearing, and we started pretending it was stil there. We have failed to recognize that we are at a technological plateau and the trees are more bare than we would like to think. That's it. That is what has gone wrong." Tyler identifies the exhaustion of easy opportunities for growth as the main culprit behind the slowdown in growth over recent decades, rising inequality, the nastiness of our present-day politics, and even the recent global financial meltdown.

It's a wonderful book: provocative, clever, full of surprises, accessibly written, and mercifully brief. And most of the constituent elements of his argument I wholeheartedly endorse. Indeed, in a weird coincidence, I have a new Kauffman paper in the works that touches on very similar themes. I agree with Tyler that growth has gotten a lot harder since the early '70s. And I agree that a great deal of political mischief has been caused by our collective failure to grasp this important truth.

But, at least for now, I can't buy into the idea of a "Great Stagnation." I just don't think the evidence supports Tyler's sweeping historical narrative that centuries of easy progress are now over. A lot of the force of his argument comes from contrasting the United States' glittering economic performance in the decades following World War II with the decidedly less impressive record in recent decades. But if you zoom out and look at the larger historical record, Tyler's Great Stagnation more or less disappears. And if you zoom in and examine recent trends in detail, the numbers likewise belie the claim that we have hit some "technological plateau."

Tyler correctly points out that median family income rose smartly after World War II only to fall off sharply in the '70s. GDP per capita figures reveal the same trend, albeit a little less dramatically (because of the rise in income inequality). Between 1950and 1973, the average annual growth rate of real GDP per capita was 2.5%; for the period between 1973 and 2007, the corresponding figure was only 1.9%

But look what happens when you put these figures in larger historical context (note: I'm using calculations by Angus Maddison for earlier periods and Census figures for post-WWII periods):

1820 - 1870            1.3%

1870 - 1913            1.8%

1913 - 1950            1.6%

1950 - 1973            2.5%

1973 2007              1.9%

From this broader perspective, what Tyler calls the Great Stagnation looks like a return to normalcy after the "Great Boom" of the post-WWII decades. Indeed, recent growth rates are better than those of all other earlier periods. So yes, growth has cooled down since the postwar "Golden Age," and that fact poses real economic and political challenges. But the Golden Age was the outlier, not our present era; it just doesn't make sense to talk about the present period as stagnant after centuries of easy growth.

Now let's focus on trends in recent decades -- in particular, productivity growth. If we've reached some kind of technological plateau, that should show up most clearly in a fall-off in the growth of output per worker hour.  But look at the actual BLS labor productivity figures for the nonfarm business sector:

1947 - 1973            2.8%

1973 - 1995            1.4%

1995 - 2007            2.7%

Once again, there was a big dropoff after the postwar boom. But then look what happened: beginning in the mid-'90s, fueled by the IT revolution, productivity growth came roaring back, nearly equaling the record of the Golden Age. It's hard to look at these figures and conclude, with Tyler, that the trees in the orchard are becoming bare.

Granted, the productivity comeback offers no ground for complacency. The productivity figures look better than the per-capita GDP figures, in large part because the labor force participation rate peaked in the late '90s, fell during the dot-com bust, and only recovered to early '90s levels by 2007 (superior growth in output per worker was thus partially cancelled out by sluggish growth in the number of workers). Meanwhile, the per-capita GDP figures look better than the median income figures cited by Tyler because of the rise in inequality -- that is, because income growth has been concentrated at the top of the socioeconomic spectrum.

These facts point to real challenges for future growth. One reason the labor force participation rate has stalled is the aging of the population -- a trend that is going to cause all kinds of economic and political headaches in coming years. And rising income inequality is due in significant part to slumping human capital formation: we reaped big gains from sending Tyler's "smart but uneducated kids" to high school, college, and grad school, but educational attainment at both the secondary and postsecondary levels has stagnated since the '70s even as the demand for highly skilled workers has continued to climb.

I've focused on criticisms here, but I want to close by reiterating what an interesting, intelligent, and thought-provoking book this is. Growth has gotten harder, and there are mounting obstacles ahead. For pointing out these sobering facts, and doing so in such an engaging manner, The Great Stagnation deserves a wide readership.

Thermodynamics and Economics

This is a terrific article by Timothy Garrett at the University of Utah, published in October: "On the coupled evolution of inflation, wealth, and atmospheric concentrations of carbon dioxide." Garrett devises a thermodynamic model of economic growth and energy use and his article is far deeper than what typically passes for energy economics analysis. Looking for a reason to stay awake at night? Look no further:

What remains is only to rapidly decouple civilization growth from CO2 emitting sources of energy. There is an important caveat however, which is that such decarbonization does not slow accumulation by as much as might be anticipated. Decarbonizing civilization promotes society [sic] wealth by alleviating the rise in atmospheric CO2 levels. But growing wealth is tied to energy consumption rates, and therefore to CO2 emissions themselves. Only a combination of extremely rapid decarbonization and civilization collapse keeps CO2 concentrations below 500 ppmv within this century. 

Why Not a Market for Advancing Innovations out of Universities?

Yesterday I had the privilege of delivering testimony on behalf of Bob Litan in front of the Joint Economic Committee. The subject of the hearing was how the United States could maintain its position on the global frontier of innovation, with a specific focus on universities and federal funding of research and development. Most of the Members of the Committee were in general agreement with the panelists: innovation is key to economic growth, universities are important to R&D and innovation, and the United States needs to make some changes if we are to remain an innovative economy.

On one particular point, however, there was a considerable degree of contention. Congresswoman Carolyn Maloney (D-NY) inquired as to the state of technology commercialization in higher education. Can it be improved? Is everything peachy-keen? 

The Kauffman Foundation takes the position that all is not well in the land of technology commercialization. There are success stories, to be sure: Gatorade, Vitamin D, Netscape, MRIs and so on. And there are star performing universities--Stanford, WARF (the Wisconsin system), and the University of California system do a phenomenal job of moving innovations out of campus buildings and into new (and existing) companies. But universities, on the whole, appear to be under-performing relative to what they could be contributing to economic growth. The Bayh-Dole Act of 1980 allowed universities to hold the intellectual property rights to innovations and discoveries that result from federally-funded research. The result was a wave of university innovations as well as a wave of attempts to move university innovations into the marketplace. Most universities receiving federal R&D dollars established technology licensing or transfer offices (TLOs or TTOs) to manage the intellectual property and serve as the liaison for faculty members seeking to commercialize their innovation. As noted, many universities have managed to successfully commercialize new technologies.

As with anything, however, the structure that was set up in response to Bayh-Dole has now run into problems. The productivity of federally-funded research at universities has been declining for a number of years. Technology transfer offices are typically understaffed and unable to manage their university's stream of innovations--innovations, moreover, that often span a wide diversity of disciplines. At the same time, however, bureaucracy has grown up around the process of technology commercialization and dramatically slowed, if not stalled, the process in many places. Anecdotal evidence abounds of faculty members running headlong into bureaucratic drywall at their licensing office. Furthermore, as a result of the outstanding success of a handful of universities with a handful of innovations, many TTOs and TLOs have implicitly or explicitly adopted a home-run mentality, chasing the next big hit at the expense of many smaller innovations. 

Finally, many universities have latched onto the most easily measured outputs of R&D, patents and licenses. But, there is little correlation between federal research funding and the returns (in patents and licenses) on that funding. Maybe we shouldn't expect there to be such a correlation, of course, but the race by universities in the wake of Bayh-Dole to establish commercialization offices and the drift of research spending toward a more political rather than scientific distribution have raised the implicit promise that there is such a correlation. The wonderful historian Roger Geiger has documented such drift: at one time, the distribution of university innovations and federal R&D money to universities both followed a Pareto distribution, with a handful of institutions far outdistancing everyone else in terms of inputs and outputs. Over time, however, while the distribution of innovative outputs remains skewed, the distribution of research dollars has gradually gotten flatter. We expect proportional output in terms of innovations but don't receive it.

There is no magic solution to these bottlenecks, but my colleagues Lesa Mitchell and Bob Litan have come up with a fantastically simple and novel idea: open up technology transfer to market competition. That is, allow a professor from University X with a potentially breakthrough innovation to go outside his or her university's TLO and use that of University Y, if University Y happens to be more skilled at commercialization in this particular discipline. University X would not be required to relinquish IP rights, and of course the professor is not required to do anything. The idea is simply to bring more openness and competition into a process that has become muddled and distorted. Where else in society do we tolerate such artificial monopolies with public research dollars at stake? This idea, in fact, perfectly carries forward and extends one of the primary motivations behind Bayh-Dole--to smooth the commercialization process. Bayh-Dole was enacted in part to construct a platform upon which faculty and professors could more easily interact with companies (new and existing). Allowing universities to hold IP rights was meant to remove one of the legal entanglements in this process.

Who could oppose such openness? Who could oppose competition when competition has been shown repeatedly to be the best avenue toward maximizing innovation and social welfare? Not to sound trite, but follow the money. Most universities and TTOs and TLOs vehemently oppose this idea. In fact, you can scroll through the webcast of yesterday's hearing to the very last question and witness the animation with which Dr. Samuel Stanley opposes the mere mention of this idea. Basically, the opposing arguments boil down to these: (a) there is no evidence that the current licensing process is dysfunctional; (b) allowing such "free agency" would give rise to all the legal problems that predated Bayh-Dole and were put to bed by that legislation; (c) no professor and no university wants to do this; (d) technology commercialization is a delicate art that cannot be commoditized; and (e) the current system is working fine as it is--as proof, look at all the great innovations universities have given us.

Unfortunately, I did not get a chance for a counter-rebuttal, so I offer it here. First, Dr. Stanley (and others) confuses an absence of evidence with evidence of absence. The absence of overt dysfunctionality does not imply the corollary, that things are hunky-dory. In fact, the arguments I've mentioned above regarding research productivity and returns to patents and licenses would seem to indicate something is amiss, as would the growing stock of anecdotes from frustrated professors and companies. It is of course nearly impossible to prove that something like a university is not living up to its perceived potential in terms of commercializing innovations, but the inability to prove that there is room for improvement is not an argument against opening up the process. That, in fact, is the nub of the free agency proposal--just open it up. No one would be forced to operate as a free agent or to contract with a free agent, as Dr. Stanley seems to think we are suggesting. This is entirely optional, as is any competitive market. It could even be optional at the level of the department or school of any particular university, with one department perhaps choosing to work with ABC agent while another chooses XYZ agent.

Dr. Stanley also predicts that such openness would instigate a return to those dark days before Bayh-Dole when legal machinations corrupted every attempt at commercialization--contract disputes, broken agreements, etc. Oddly, however, technology commercialization at universities is still plagued by these problems--Bayh-Dole did not banish legal disputes, and I'm not sure how Dr. Stanley manages to pull off such cognitive dissonance.

By opening up the technology commercialization process, we would likely see the development of completely new organizational forms to facilitate the process. No one can predict what will happen--and that is exactly the point. Maybe free agency will flop; maybe it won't. But there is no (or shouldn't be) single model when it comes to technology commercialization: free agency would encourage experimentation as a way to lower transaction costs, which was precisely the idea behind Bayh-Dole. In fact, many professors are already finding their way around the TLOs and TTOs, doing "end-runs" around the bureaucracy and finding their own commercialization routes. Rather than see this perversion as a defense of the status quo we should encourage free agency as a way to promote the development of more such alternate pathways. In their promotional materials and commencement speeches, university administrators consistently trumpet the glories of diversity and self-actualization and intellectual experimentation--except, it seems, when it comes to their own operations.

That seems to be a running theme of opposition: technology commercialization is special and cannot be subject to the rules of the market that govern the rest of the economy. Dr. Stanley is far from alone in asserting that each particular innovation is unique and must be granted its own customized commercialization process that can only be run through a TLO. Undoubtedly, there is some truth to this. Just as clearly, however, there is plenty of room to lower transaction costs by standardizing some portions of the commercialization process. This is precisely what UNC-Chapel Hill has done with the Carolina Express License Agreement; we will keep you posted as to whether or not the world ends as a result of this dangerous experimentation with technology commercialization.

Finally, we come to the tried and true defense that everything is fine. Universities have, after all, contributed enormously to economic growth and human welfare. Why mess with success? The real message, however, is: don't mess with us! This is the completely predictable and understandable response of any institution when its monopoly is threatened and should, in fact, serve as a reason to promote more experimentation. Just because universities have contributed much in the past does not mean they are performing optimally today (research is only one of many fronts on which we could have this discussion) and does not guarantee that they will continue to do so in the future. If bureaucracy keeps building up around the discovery and innovation process, it is almost guaranteed that they will not.

From "Brain Gain" to Brain Circulation

[Cross-posted from Progressive Fix]

A recent post highlighted the importance of new and young companies to job creation in the U.S., implicitly raising an important question for policy makers: How can we increase the number of startups? Assuming it can be done, such an increase would not solve all of the economic challenges facing this country, but it would certainly help. New companies not only create millions of jobs across all sectors of the economy — they also introduce product and process innovations, boosting overall productivity.

Saying startups are important is one thing, of course; actually designing policies to increase their number is something else entirely. Before making any recommendations, for example, we need to know more about the universe of startups. Are they more prominent in some sectors than others? Does the impact of new companies differ across sectors or geographic regions? Should policy focus on encouraging more new firms, or on enhancing the growth of those already in existence? How would any such policies affect established companies, large and small?

Policymaking around entrepreneurship is evidently not clear-cut as there is still quite a bit we do not understand regarding startups. In the coming weeks we will try to explore these questions and illuminate the world of startups for policymakers. We’ll start with the lowest-hanging fruit of all, though one that may seem like poison to some in Washington: immigration.

It’s commonly accepted that the United States is a nation of immigrants, settled and populated by those fleeing persecution, seeking commercial opportunities in a new land or looking for a fresh start. We have always recognized the important contributions of immigrants to the U.S. economy, from entrepreneurs like Samuel Slater (textile mills) to Andrew Carnegie (steel) to Andy Bechtolsheim (Sun Microsystems) to the laborers and workers who built this country with their hands.

Recently, researchers have begun to paint a broader picture of the economic role of immigrant entrepreneurs. For example, Vivek Wadhwa and his research team have found that, from 1995 to 2006, fully one-quarter of new technology and engineering companies in the U.S. were founded by immigrants. In Silicon Valley, the figure was one-half. These firms constitute only a sliver of all companies, yet contribute an outstanding number of jobs and innovations to the economy.

It makes sense, then, that if we are seeking to increase the number of new companies started each year in the U.S., we might look to immigrants. It turns out that Sens. John Kerry (D-MA) and Richard Lugar (R-IN) are thinking precisely along these lines, introducing the StartUp Visa Act (PDF) in the Senate. This bill would grant a two-year visa to immigrant entrepreneurs who are able to raise $250,000 from an American investor and can create at least five jobs in two years. Without question, such a visa is a good idea and this legislation hopefully paves the way for future actions that would reduce the pecuniary threshold and focus more on job creation.

Quite naturally, however, the promotion of immigrant entrepreneurs arouses suspicion among those on the right who harbor nativist views, and those on the left who perceive progressive immigration policies as a threat to American labor. Such views take the precisely wrong perspective: immigration, as we have seen, is a core American value. Immigrant entrepreneurs, moreover, come to the U.S. to make jobs for Americans, not take them.

Further, many of those who promote immigration as a way to boost economic growth narrowly focus on “high-skilled” entrepreneurs, those who might start technology companies. Clearly, as Wadhwa’s research indicates, such companies are important to American innovation. But we exclude non-technology entrepreneurs at our peril — every new company, including those founded by immigrants, represents pursuit of the American dream. By closing our borders to immigrants in general or welcoming only those with certain skills, we leave out many who will start new firms in other industries. If not in the United States, they will go elsewhere to start their companies and create jobs.

Entrepreneurs are implicit in Emma Lazarus’ poem: “Give me your tired, your poor/Your huddled masses yearning to breathe free.” Entrepreneurs start from nothing and work endlessly to build their companies, expressing their individual freedom through commerce. Why should we want to exclude them from the home of entrepreneurial capitalism?

ADDENDUM: Ben Wildavsky's new book, The Great Brain Race, will be released soon and I highly recommend it to anyone interested in higher education, globalization, and immigration. It is a fantastically written book and, as noted in the title of this post, Ben posits this wonderful notion of "brain circulation." When discussing globalization and immigration and job creation in the very recent past (and, for that matter, still in the present), we have usually spoken in terms like "brain gain" and "brain drain," implying a zero-sum contest between countries and regions. Ben's book makes quite clear that we are moving into a world of brain circulation, wherein people circulate among countries and institutions, starting companies, creating jobs, propagating innovations--adding to the economies of many countries at once. 

The Accumulation of Firms and the Predominance of New and Young Companies

This is a sequel to last week's post about moving from Paul's drunkard's walk to the accumulation of firms over time. Basically, the message of those two posts is that the internal dynamics of the American economy--lots of new companies founded each year, the exit of firms at all ages, the persistent predominance of young firms as the economy moves through time--make it highly probable that job creation will come from new and young companies rather than older, more established companies. (In these data that we're using, a firm is no longer classified as "young" once it reaches age 6, so the term "older," perhaps imprecisely, includes firms ages 6 and up.) Thus, when we say that new and young firms account for most net job creation in the U.S. economy, we are partly describing the reality of these firm dynamics.

This doesn't diminish the importance of these companies or the jobs they create. After all, the reality of these dynamics is premised on four things: a more or less steady inflow of startups; more or less consistent survival rates from year to year; comparable gross job creation and gross job destruction in older companies; and a higher number of jobs created at young firms (4 versus 1). These are not functions of what might be perceived as mathematical inevitability--they are the reasons behind the mathematical probabilities we are describing. If the level of firm formation were more volatile, if older companies created more jobs than they destroyed, if young firms were somehow suppressed in their hiring (through, perhaps, bailouts of larger, older companies; just speculating here)--if these variables changed, the dynamics of firm age through time, and subsequent job creation, would be different.

But, in recent history, these variables have all held and together they have created a picture that does, in fact, resemble something like mathematical inevitability, but which is probably more accurately described as a likelihood, or something of higher probability. In any given year over the past three decades, young firms (startups [age 0] and firms ages 1-5) have constituted the largest demographic of firms in the United States. Starting an economy at t=0, going out 25 years, the influx of new firms and the survival rates of firms over time generate this outcome. And, we might expect (provided the other variables are true) that the largest demographic sector will create the largest number of new jobs. This says nothing, of course, of innovation or productivity, subjects for another day.

Here, I want to supplement last week's post, which used average figures calculated from the Business Dynamics Statistics data, with the actual numbers from that dataset, which covers 1977-2005. That is, did reality conform to the stylized numbers we used last week? (The charts will not follow precisely the same format because last week's charts extrapolated out beyond 5 years on an annual basis, while the actual data aggregates firms into 5-year age categories after the fifth year: 6-10, 11-15, and so on.) We'll start, first, with the U.S. economy starting de novo in 1977, with only firms that came into existence that year. How, over the next 28 years, did the composition of firms by age change?

Here, startups (age 0, new firms founded each year) are broken out separately while all other firms are combined into 5-year categories. The data generally track the accumulation pattern revealed in last week's post: in any given year, young firms (ages 1-5) will quite naturally comprise the largest sector in terms of firm demographics. Note how few firms in 2005 are included in the 26-28 age category--these firms were founded in 1977-1979. By 2005, 19% of the firms founded in these three years remained. That may be more than expected, but using the survival line for firms founded for 1977 we can trace a slowly declining line of survival out two decades. If we imagine that the line continues on that downward slope for the next two decades, it appears as if very few firms in the economy survive more than four decades. Either those firms we all delight in describing as "dinosaurs" should properly be seen as remarkable achievements, or firm failure (over time) is a much more dominant force in the economy than we perhaps imagined. Probably both, and I'm probably simplifying here.

The advantage of using the real data from BDS is that we can include the "left censored" firms in our calculations, those companies founded prior to 1977 but for which we don't have age information. In 1977, then, these firms could be anywhere from age 1 to age 100. But we can track their survival after 1977 against all firms founded after that year. How long do these companies persist?

I included a brief explanation within the chart itself, but you can clearly see how these pre-1977 firms gradually decline in number over time, while new and young firms in any given year gradually get bigger in comparison. By 1988 (the circle), the pre-1977 firm group has shrunk such that it is now smaller each subsequent year compared to cohorts of new and young companies. By 2005, these firms still number 1.5 million, accounting for about one-fifth of all firms in the economy. This is a lot, but is an aggregate of every year before 1977, so there are probably a good number from the 1970s, a proportionately smaller number from the 1960s, and so on. 

Now, we can include these pre-1977 firms in our overall chart of firm demographics over time and compare firms of all ages:

This is actually an incredibly interesting chart, as the text insert fleshes out, but I'll repeat it here so you don't have to open the image itself. It shows two different ways in which firm composition changes and in which new and young firms consistently account for the bulk of firms in the economy. In 1977, the t=0 moment here, the economy is composed  of a class of startups (age 0) and all other firms existing that were founded before 1977. As the economy moves through time, you see four things happen. First, age 0 firms and age 1-5 firms account for a steady slice. Second, the overall number of firms accumulates. Third, the pre-1977 firms gradually decline in number. Fourth, post-1977 firms that age through time also diminish in terms of their numbers as fewer survive in subsequent years. So new and young firms dominate the economy (numerically) by two mechanisms: gradual disappearance of older firms and gradual disappearance of firms as they age. Same thing, different pathways (on this time scale).

Finally, we'll move to the percentage shares of each firm demographic (excluding pre-1977 firms here):

Once we get to a point where we have firms older than age 6 (and remember, the pre-1977 firms account for the missing shares here) in 1983, new and young companies account for between 30 and 40 percent of all firms in the economy. This percentage naturally falls over time, albeit slowly, as the overall number of firms grows, but this demographic of firms nonetheless comprises the largest demographic slice--a plurality, if you will--in each year, and this would persist over time.

So what, if anything, does this all mean? It means that new and young companies account for most new jobs in the economy in part because they are the largest demographic category of companies. It also means that job creation may actually be a secondary story here because this reality is a reflection of underlying dynamics that have been remarkably persistent over the past three decades. The United States has enjoyed a more or less steady level (and rate) of firm formation and survival rates for those firms have also been rather steady. Older companies, moreover, experience roughly equal levels of gross job creation and destruction, meaning their net job creation approaches zero.

It is entirely possible that these are transitory phenomena. As far as I can tell from the desultory data and research on firm formation prior to the 1970s, entrepreneurship (and subsequent survival rates) across the entire twentieth century have been remarkably consistent (I mention this, with an accompanying citation, in a forthcoming paper, so I'm not simply making this up). According to some Census data, for example, the number of new business created each year doubled from the 1940s to the 1990s--yet the country's population also doubled so the per capita rate of firm formation remained steady. Over the past thirty years, moreover, entrepreneurship rates have been remarkably steady, not declining as some have alleged. 

Yet we surely can't assume that the dynamics beneath such factors are immutable. The preponderance of research concerning firm demographics in many European countries, for example, holds that firm exit is a much smaller factor and that, consequently, unproductive firms simply continue to exist and new firms are dissuaded from entering. Large companies enjoy favorable policy and so the above charts might look quite different for other countries and clearly indicate that the dynamics can change. Even if part of what the above charts do is describe a mathematical reality, it could be considered progress if policymakers in Washington and elsewhere spoke in terms of this reality rather than giving undue attention to the largest companies in the economy (which, incidentally, often seek regulation as a means of protection from competition from startups). What this all means is that startups and young companies are no less important than we thought before--they are a robust and enduring dimension of the U.S. economy. It would be foolish, however, to assume that this is somehow a natural or permanent state of affairs--policy could easily be made to alter these dynamics for the worse.

Startups in an Evolutionary Model

In describing how human knowledge advances, Donald Campbell devised the BVSR model to explain the evolutionary way in which knowledge accumulates. BVSR stands for Blind Variation and Selective Retention and is more or less natural selection applied to creativity and epistemology. Others have explored Campbell's with great insight and how it applies to fields such as organization science.

Evolutionary economics moves in and out of fashion but for the most part remains on the fringes of economic thought. Prompted in part by Paul's post earlier today, I'm wondering today about startup firms in a BVSR model of economic dynamics. It's no surprise (and, I don't think, controversial) that free market competition works through an evolutionary process of selection. Individuals and companies try various things (new products, services, processes) and their fate hinges on whether or not the market "selects" them vis-a-vis competing ideas. These new products and services are the variations in BVSR, whether radically new innovations or incremental improvements on something that already exists. We throw lots of spaghetti against the wall to see what sticks. Economic variation, of course, is not completely random since firms often seek to satisfy a previously identified market need or solve a heretofore unsolved problem. Variation is, however, blind in Campbell's sense, in that we do not know beforehand what will or will not succeed. "Blind" refers to our knowledge (more specifically, our lack of knowledge) about the consequences of any particular variation.

So we continuously introduce new variations into the market. The manner in which they are winnowed and propagated is "selective retention." Selection is clear: a new market variation either succeeds or it doesn't. Or, more precisely, it either survives (is selected) or doesn't survive (is not selected or selected against), with survival not always a proxy for subjective superiority. Here, it seems like the BVSR model advances somewhat beyond the standard view of the free-market-as-natural-selection (in addition to the explicit recognition that variations are blind). Once a variation is selected, what happens to it? It is retained, meaning it somehow propagates through the population. A variation doesn't merely survive--it grows and spreads. 

How would you increase economic growth or increase innovation through the BVSR model? Let's say each element is multiplied by the others, so growth or innovation is the result of V x S x R, all divided by B. Increasing any one element in the numerator should theoretically increase the product. (It's not clear that this should be true, by the way. I haven't convinced myself that multiplication is correct here. I suppose it's conceivable that it could or should be, for example, "+" R. This isn't a research paper and I'm just following an idea here, so please feel free to take issue with this. At this point, it's of de minimis concern.) According to this, then, you could increase growth or innovation by increasing variation (the introduction of new products and services), improving the selection process (so only "good" or "right" variations are selected), and improving retention, to the properly selected variations quickly achieve rapid scale and propagation, disseminating their benefits to everyone.

How would we approach startups here? Obviously, startups are one type of variation (at the population-level of the economy). They are selected for and against: roughly half of new firms (48%) survive to age five, with the survival curve slowly falling after that. (I've seen some data indicating that one-third of new firms survive to either age 7 or age 10.) Survival, of course, tells you little about success or quality, and selection in the case of firms is an ongoing process. Some research suggests that only the tiniest handful of companies survive for more than a few decades (suggesting that groups such as those companies on the Fortune 500 are both an incredibly small minority and, perhaps, a secondary story--maybe failure is the dominant story of economic selection). Retention could be either the organic growth of young companies or, I suppose, acquisition by larger, more established companies. (Revenue or employment growth or both would here serve as a proxy for another way of looking at retention: spread of a company's product through the market.) With the denominator, "Blind," we might want to reduce the blindness of our variations--that is, wouldn't we want to have a better idea of the prospective fate of new firms? Or, should B be another multiple here? Would increasing blindness be better because it would throw more of the burden onto the competitive process? Perhaps B is neutral here? That is, perhaps it cannot be increased or decreased? 

In any case, let's assume, with good reason, that startups are absolutely essential to innovation and economic growth. To increase these, what would we do with startups in the BVSR model? To begin with, could we reduce the blindness of a particular startup's fate, or the fate of all startups together? (It's not clear, moreover, what Blind should refer to here: mere survival to a certain age? Success, however nebulously defined?) Let's say Blind refers to the state of knowledge facing the founders of a new company: they do not know, and can't know, whether or not they will "succeed," however they define that term. (Although, if your objective is to simply work for yourself or the ability to wear flip-flops to work, then merely starting your own company can be considered success.) From a macro perspective, it's not clear that we can either reduce or increase blindness at the startup stage--the figure given above for survival to age five (48%) has been remarkably consistent across time. The survival rate for new firms, that is, doesn't appear to vary much founding year. There is some variance by sector but again, the survival lines within each sector are rather consistent, so it's not clear how we could either raise or lower the survival line. Of course, if survival rates have been consistent for thirty years, then perhaps the Blind element is neutralized: there is a 50-50 chance your firm will still be around five years from now.

Now, to Variation. Here, you would simply want to increase the number of new firms that are founded each year, with the expectation that trying more new variations would lead to more "good" things being selected and retained in the economy. An increase here means an increase there. I am sympathetic to this strategy, but I'm unclear as to how we do it. Firm formation in the United States has been pretty steady for two or three decades and the question of how we increase it is not so simple. Things like awareness and lowering barriers to entry are probably our best bets but these will vary by sector and geography and characteristics of the founder. All of this assumes, of course, that a basic increase in startups (more variation) will result in an increase in innovation and growth. And this assumes that we would see a similar proportion of innovative ideas and potentially successful firms in a larger pool of variations--is that a valid assumption? 

Selection of startups occurs through a variety of mechanisms. One, of course, is the market itself: you start a new firm and quickly find out whether or not it will selected in or out. If we increased the number of startups (increased V), then conceivably our selection mechanisms could remain the same, simply working on a larger population of firms. If, however, V remained constant, we might want to intensify S by looking more carefully (somehow) for promising ideas or "sure-thing" innovations. I suppose this is the role played by venture capital firms and accelerators such as Y Combinator. They have many entrepreneurs approach them and apply to them, but only select a few for funding or developing. Those that they do not select still go on to start new companies, so the claim of these particular selection mechanisms is that they are intensifying S so as to increase R: selecting the "best" of V, those with the most potential for retention and propagation by the market. As an overall strategy to enhance Selection, we would look to increase market competition in general, removing protective barriers around certain industries and firms, for example. A perennial complaint about some European economies (and conclusion of numerous McKinsey reports) is that they are not as competitive as the United States--stultified selection suppresses innovation and economic growth. It's also conceivable that simply increasing V would naturally increase S (they are, after all, multiplied by each other)--more firms overall would mean more firms competing. 

These notions of Selection, however, presume that any selection process, whether in biological evolution or in free market competition, is an optimizing one--that the "best" organism or population or idea or company always beats out inferior competitors. Adaptive fit, after all, implies such optimization. Yet we know that natural selection is only one way in which genes and organisms are winnowed and propagated. The role of accident and randomness in evolution cannot be forgotten: we like to think of dinosaurs as having met their proper evolutionary fate because they were all slow and dim-witted. But many evolutionary theorists have pointed out that, if a meteor or volcanic explosion did the dinosaurs in, then that isn't necessarily "natural selection" in terms of adaptation and competition. It was an exogenous event and the emerging population of smaller mammals happened to be better suited for survival. So, yes, an exogenous event "selected" out dinosaurs, but it wasn't necessarily the "optimizing" path we like to imagine. The same applies to the selection process in terms of firms and ideas: Silicon Valley bested Boston and Dallas because it had a better ecosystem, the selection process was more optimizing. But what if William Shockley's mother lived in Dallas, not Palo Alto? What if Frederick Terman had not contracted tuberculosis and had ended up returning to MIT? I realize these are entirely speculative and that the obvious riposte is, 'Well, that's just how things happened.' My point is simply that we shouldn't assume that Selection always implies optimizing; human behavior and group dynamics are such that the economic process contains plenty of satisficing behavior. (We also shouldn't forget that an entity such as a firm can become over-adapted to certain circumstances and thus too sessile to change; selection could work too well.)

Finally, how would we enhance Retention? What would this mean? It could entail somehow promoting those firms that are selected--say, those that survive to age six--via certain rewards such as a retroactive tax credit or something like that. Or, it could mean favoring those firms that grow rapidly in their first few years and have rapidly passed through V and S and we want to enhance their retention and propagation. It could mean reducing barriers to "scale," whether in the form of regulations, corporate taxes, whatever. (I continue to maintain, however, that the effect of certain taxes on entrepreneurship is not nearly as straightforward as some might like to believe.) Maybe we think of Retention as going public, in which case perhaps changes need to be made to Sarbanes-Oxley. Over the past few years, around 80% of VC exits have been acquisitions, not IPOs, and many of the companies coming out of accelerators are acquired by companies such as Google.

In any case, I don't know if this has added anything to the general discussion of entrepreneurship and economic growth or not. If we think about startups through the BVSR model, there are probably certain areas in which we can make changes to enhance the economic impact of startups, although some of them steer quite close to "picking winners." In general, as I said, I favor increasing V, but I'm as yet unclear as to how we go about doing this. Importantly, increased Variation should apply to firms of all types, not simply in favored sectors or those sectors we think (or would like to think) represent promising areas of innovation. Firm formation and innovation are compounding phenomena, so increasing the overall level of startups should (emphasize, should) raise our rates of innovation and growth, and this should be true whatever sector of the economy we are discussing. Information technology has been hugely important for American productivity, both in its production and its use in sectors like retail. We needed IT, but we also needed all those new retail outlets and supply chain innovations for the IT to become productive. It's possible that a BVSR-type of analysis in, say, 1995 might have concluded: we need more IT startups and fewer new retail stores. The economy might not have performed as well as it did.

More on Friedman

Here is another, albeit similar, take on one particular response to Thomas Friedman's Easter column celebrating startup companies. By the way, while you visit the PPI website, be sure to read Michael Mandel's policy memo on why the jobs crisis may well be an innovation crisis. 

Defending Thomas Friedman on Job Creation and Startups

In his column yesterday, New York Times columnist Thomas Friedman quoted Kauffman's Bob Litan about the importance of new and young firms to job creation and emphasized the importance of immigrants to firm formation. It was a trademark Friedman column: factoid, examples, quotations from experts, and concise policy lessons. I'm biased because Friedman used Kauffman data, but it was a good column.

Not everyone, however, wants to hear about the economic change that new firms and immigrants visit upon the U.S. economy. At Beat the Press, Dean Baker took issue with Friedman's column and Kauffman's data. With a layer of snark that is by now wearily standard in the blogosphere, Baker writes: "Friedman's conclusion about the special importance of new firms is utter nonsense." His logic:

The claim that most net new jobs came from new firms conceals the fact that existing firms added tens of millions of jobs in this 25-year period. Of course existing firms also lost tens of millions of jobs. We can say that the net job creation for existing firms was zero, but if we did not have an environment that was conducive for the job adders to grow (how many jobs did Microsoft, Apple, and Intel create after their first 5 years of existence?), then existing firms would have lost tens of millions more jobs.

Well, yes, that's true--and that's precisely the point. There are basically two ways to look at job creation in the economy: gross and net. On gross terms, large existing companies hire thousands of people each year, but they also see thousands of people leave, voluntarily or involuntarily. Gross job inflows and outflows in the American economy are enormous, an indicator of the ongoing and dynamic reallocation of resources that drives economic growth.

At the end of the day, however, if we want to keep pace with an expanding labor force (new entrants) and changing economy (the rise and fall of sectors and companies), what matters is net job creation. We have fifteen million people unemployed in the United States right now, and over the next several years net job creation has to proceed at a remarkable pace to not only provide jobs for this population but also create jobs for the millions of new people who will enter the labor force in the coming years. The last two years have seen the two biggest matriculating classes ever in American higher education, which means that in four or five years we will see a large influx of new graduates searching for jobs. How does Baker propose we keep pace with such growth?

From a macro perspective, for keeping pace with this growth, the only thing that matters is net job creation. It would be little consolation if we had 100 people looking for jobs, and large company ABC hired those 100 people but also fired 100 different people. For Baker, this is a good thing. And it is--for the 100 people who were hired. But what does Baker do with the 100 newly unemployed folks? Relate to them the wonders of big companies and their gross hiring?

Big, established companies are indisputably important--the symbiosis between new and old, small and large is a key strength of the U.S. economy. But we only get net job creation--a positive increment of new jobs to keep pace with labor force growth--from new and young companies. True, this population is highly dynamic, equally adept at creating jobs as eliminating them. Whereas this dynamic process works out to a net gain of zero in big companies, the net result among new and young companies is positive. If you subtract startups from the American economy over the past thirty years, there is only a handful of years in which the economy would have created a positive net number of new jobs. This makes sense as a brand new company can only create, not destroy jobs, but from the perspective of adding new jobs--for those 100 people above--this doesn't change the fact that new jobs come from new and young companies.

Baker also neglects to mention that once a company survives to a certain age (at age six a company is no longer classified in the data as "young") it becomes not an organic generator of jobs but an acquirer of new jobs via other companies, which are largely young and small. Baker asks, seemingly rhetorically, how many jobs Microsoft added over the age of five, but companies like Microsoft and Intel and eBay and others can generally only add a positive increment of net new jobs by acquiring young companies. So even the appearance of net job creation in our largest companies reflects the contribution of new and young firms. None of this, moreover, even begins to approach the role of new and young companies in innovation and economic expansion--not monolithic growth but the evolution of sectors and firms into new and unknown territory.

Baker also appears to cast cold water on Friedman's emphasis on the importance of immigrants, but doesn't dwell on it. There is, however, a faintly palpable nativist tinge in the comments to Baker's post. Why bring in "foreigners" to start companies? Don't we trust Americans? Nowhere, however, does Friedman (or Kauffman) claim that Americans are insufficient for entrepreneurship. In fact, immigrants who start companies in the United States create thousands of jobs for Americans--who could argue against that? For a country made great on a history of immigration, the United States risks falling prey to nativist sentiment from both the right and the left, which will inevitably stifle economic dynamism.

In criticizing Friedman (and Kauffman), Baker tries to cover his tracks with snark, wishing someone would revoke Friedman's license to discuss economics. In an altogether brilliant post last week, Michael Pettis observed, "Snarkiness is always a sign of intellectual insecurity or confusion, and never an indication of insight." Rather than snark, Baker should consider how his vision of a big-companies-only economy would lead to stultified job growth, harming the people he claims to want to protect.

How India Changed Growth

Growth economics is filled with errors. The Solow residual has become ingrained in conventional wisdom as "technology" when it is really a measure of intangibles. Technology, sure, but also culture, institutions, creativity. Not to mention, conventional wisdom assumes a linear link between more R&D spending and more innovation. What about entrepreneurs?  Most startups I know don't have an R&D budget -- the whole company is a tech gamble!  How can government officials measure that?  They can't.

But the biggest myth of practical growth economics is getting punctured, thanks to India. The CW still hasn't digested the truth yet, because it is still beholden to the idea that manufacturing is the alpha and omega of prosperity.  Well, India has been showing that a country need not "industrialize" to grow. Rather, India is leap-frogging (for the most part) the trail of sweatshop tears by embracing services-led development.  You might call it a Globalization strategy instead of Industrialization strategy. And lest we forget, embracing Industrialization was no guarantee of success. Remember Africa circa 1960?

Here's Ejaz Ghani in a VOX essay:

... India’s growth pattern in the 21st century is remarkable because it contradicts a seemingly iron law of development that has held true for almost two hundred years since the start of the Industrial Revolution (Chenery 1960, Kaldor 1966). This law – which is now conventional wisdom – says that industrialisation is the only route to rapid economic development for developing countries. It goes further to say that as a result of globalisation the pace of development can be explosive. But the potential for explosive growth is distinctive to manufacturing only (UNIDO 2009). This is no longer the case.

... Service-led growth is sustainable because the globalisation of services is just the tip of the iceberg (Blinder 2006). Services are the largest sector in the world, accounting for more than 70% of global output. The service revolution has altered the characteristics of services. Services can now be produced and exported at low cost (Bhagwati 1984). The old idea of services being non-transportable, non-tradable, and non-scalable no longer holds for a range of modern impersonal services. Developing countries can sustain service-led growth as there is a huge room for catch up and convergence.

Maintream economics taught students that services dominate the non-tradeable sector (think haircuts) across distance.  I always wondered what Elvis or Lassie would say to that.  But now I just wonder what the accountants in India would say.  Probably the same.  Woof. 

This is worth keeping in mind the next time someone tells you that "America used to be a place where people made things."  It still is, but that's not the future. In the future, manufacturing will be automated. American factories will quietly produced more stuff with less, or no, workers on the line. Indeed, that trend may make China's industrialization strategy a bad bet over the long-term.

Paul Romer's Errors

This is an overdue summary of the presentation Stanford's Paul Romer gave at the signature growth panel at the Atlanta AEAs ...

Romer started by observing the divergence of international incomes has been a powerful and vexing trend. Despite the ever-widening gap between developed and developing economies, he also noted research showing how life expectancy across countries has converged. This bodes well for the future, and suggests to me a stepping stone to convergence in other things.

Second, Romer made a fantastic point that the workhorse model of gains from trade (globalization) is comparative advantage, but it is a story that should be updated with new goods. Instead of the two trade goods being fish and apples, for example, we should talk about trade in cholesterol pills and blood pressure pills. The first implication seems to be the consumption benefits of trade are too often neglected. The implication Romer was making is that trading in recipe-driven goods represents an inefficiency. We should imagine a globalized economy where production is localized and IP is licensed, rather than one where final goods (physical objects) are traded.  In his words, "globalization should be a flow of ideas, not goods."

But the real insight offered in Romer's talk was when he called out the "two errors" of development thinking. He got to them by sketching out a few growth equations, starting with the familiar 

Y = A * F(K,H,L)

and then expanding the A term as a function of technology and rules:

Y = A(Technology,Rules) * F(K,H,L).

Understanding the importance of rules is the key insight. Private property is a kind of formal rule governing social interaction. We have other rules for marriage, educational advancement, and so on. Technology is the conventional way economists thought about the growth equation and includes hardware like computers, phones, and printing presses.

When thinking about how to accelerate economic development, the first error people tend to make “Technology cannot change.”  The given tech level of a country is given. The second, and more important, error is that "Rules can change with stroke of a pen.” While there is a growing consensus that rules are the most important factor in permanent changes in a developing country, Romer forces us to accept that rules are very difficult to change.  Nations in particular, even when its leaders recognize the need for rules to change, have difficulty making them happen.

Here in America, there is tremendous acrimony about new legislation of almost any kind. Witness the current fight in the U.S. Congress over health care reform. So why do we imagine it would be trivial for another country to change its laws, let alone norms, customs, cultures?  So the real challenge for development practitioners is how to make rule-changing easier.  Any thoughts?

Is Economic Complexity Fragile?

In response to Dane's fine post on the rising complexity and fragility of an advanced economy, I have to wonder if the original essay by historian Bryan Ward-Perkins (see Financial Times op-ed) is confusing complexity with scale.

Complexity is a fascinating field, and I think the robust comments the post received are testimony to that (superb discussion!). Also, I adore the folks and research at the Sante Fe Institute, the capital of complexity studies, But the word "complexity" has become woefully trendy, and calls to mind the overuse and abuse of otherwise interesting topics like dolphins or innovation or Eastern religions. And in this case, I think it has been misapplied.

Let me quote both Ward-Perkins followed by Dane here:

"The more complex an economy is, the more fragile it is, and the more cataclysmic its disintegration can be."

Ward-Perkins' main point that complexity increases fragility, is wrong.

Keep in mind that Ward-Perkins is analogizing the current U.S. economy and recession to the British fringe of the Roman empire after the sack of Rome in 410. Rome v America is an interesting comparison, but Ward-Perkins' analysis is shockingly off target in linking 410 to 2010.  I say shockingly because his book on the collapse of Rome is perhaps the single best accounting of hard evidence on Rome's demise that I've come across.

If you are interested in the fate of large nations, you study Rome. And if you study Rome, you need the facts, which is what Ward-Perkins does better than anyone.  Fact number one is that the collapse of Roman civilization was real and terrible. His book destroys the trendy revisionism that suggests the post-Roman world was a gentle evolution and that no "civilization" was lost. This cautionary tale can not be overemphasized because it must be understood that systemic collapse could indeed happen again. Unfortunately, my admiration for Ward-Perkins' masterful documentation of the effects of Roman collapse do not extend to his attempt at causes.

Fact number two is that the collapse of Rome was not sudden. Ward-Perkins knows this, so why make the complexity argument?  It sounds good sure, but it doesn't fit. The demise of scale economies is better understood as a consequence, not a casue, of collapse. Besides, the basic causes of Rome's weakend economy are not so mysterious. Bruce Bartlett wrote an excellent essay on this years ago (worth digging up). But the key point is that bald economic ignorance by its rulers set up Roman civilization for epic failure. It wasn't their fault, in one sense, because nobody knew economics then. Nonetheless, the city fell in 410, but that was preceded by more than a century of ... let's just list the key factors ... violent political strife, hyperinflation and currency debasement, regulated labor market rigidity in the extreme, slavery and the exhaustion of its supply, and the ultimate walling off of the outside. Any one of these is a better explanation for unsustainability and collapse than "fragile complexity."

Rome, like China, ultimately walled itself off from the world, became insular, became stagnant. talk about a non-entrepreneurial economy!

Indeed, the one idea I would like to push into the minds of modern policymakers is that economic growth relies precisely on the concept of scale. With scale comes labor specialization. This was the insight of Adam Smith (and an insight the Romans lacked).  Scale can be increased externally by openness to other economies -- through trade, immigration, investment, and ideas. But scale can also be enhanced internally. Smart pro-growth policy recognizes both. Internal scale rises with bigger, better transportation networks. Scale rises with better communications technology. Indeed, the revolution in scale spawned by the internet and other digital technologies is the key reason to have confidence in a coming long boom.

Is economic scale fragile?  I suppose it can be, but that is a much deeper question. Networks, as Dane describes, can be strong or weak. A strong network (think cellular telephony) can lose one or more nodes and still function. A weak network (think of local plumbing) needs all of its nodes to be up for the system to function. Overall, the modern economy leans much more towards the strong network. Consider, for example, what happens to all the other cities in the world economy when one is incapacitated. Not much. As tragic as Hurricane Katrina was on the lives of millions of Americans, the economic lesson was a surprise. The national economy was robust in the extreme -- economic growth barely flinched at the loss of a major port of trade, even the destruction of huge amounts of oil refining.

Checking in from the AEA conference

Here we are in Atlanta, host of this year's biggest annual meeting of economists from around the world, the AEAs. I had the fortune of bumping into David Warsh at the beginning of the one star-studded panel on growth, and we sat together near the back to list to Paul Romer, Daron Acemoglu, Dani Rodrik, and Philippe Aghion discuss "Growth in a Partially De-Globalized World."

Each presentation was interesting, but somewhat narrow I thought, aside from Romer.  Romer is a big thinker's big thinker, and his concept of a Charter City can, in my opinion, truly change the world. Overall, the lack of panels on growth is a bit disappointing. Certainly, a great deal of attention is focused on macro and basic discussions of the near-term prospects for the U.S. and world economy. But I get the sense that growth is a bit stagnant again, in a kind of rut that it was almost 20 years ago before endogeneity shook things up.

The full listing of panels and some of the papers available for download are at this link:
http://www.aeaweb.org/aea/conference/program/2010_AEA_meeting_papers.php

Not to slight Acemoglu (his co-authored paper is a cautionary note about the possible negative correlation between rising militarism and declining trade openness -- a first stab that seemed somewhat obvious while at the same time historically myopic ... was WW1 really about the "end" of globalization or rather just a pause?) or Rodrik (his presentation focused on China and the dilemma of its currency manipulation, pretty interesting but I honestly didn't learn anything that changed my worldview, not that it wasn't impressive) or Aghion (his heterdox call for a smart industrial policy did not go over well, but his comments on other presentations were solid), but Paul Romer's presentation really merits extensive notes.  I'll post those next...

Field of Dreams is an Awesome Baseball Movie, not an Economic Development Strategy

By some estimates, the United States currently has vacant office space running into the hundreds of millions of square feet: law offices, Fortune 500 companies, cookie-cutter glass boxes constructed on sandpiles of debt, etc. And, as discussed previously, commercial real estate (CRE) could very well be the next economic shoe to drop, dragging down whatever economic recovery is materializing. 

At the same time, it's increasingly clear to more and more people that recovery and expansion simply won't happen without entrepreneurs: new companies that bring innovations to market. Entrepreneurs are not the silver bullet solution, yet growth will not occur without them. 

So let's see: how could we possibly address a looming CRE problem as well as encouraging incipient and potential entrepreneurs? Got it! Construct completely new science parks! Ignore both problems and travel a well-trod path that, if not inspired by Field of Dreams, at least follows the same blueprint: Build it and they will come. 

Oh, wait, hold on a second, I'm being told that this idea is already being pursued--it seems that Congress has taken time from its time-consuming BCS debates to pass legislation that will ostensibly "spur innovation [and] economic with development" by building science parks. And wouldn't you know it, the bill miraculously finds empirical support from the Association of University Research Parks. This is a page right out of the NVCA's playbook.

Not that science parks are necessarily bad, but evidence on their effectiveness appears mixed and, in any case, the concept would seem to contradict, you know, the entire history of innovation.

The bill's title is "The Building a Stronger America Act," which should not be confused with Build a Stronger America, an entrepreneurs' movement that seeks, inter alia, to expand the popular notions of entrepreneurship and innovation beyond science parks and venture capital, beyond the idea of walling off our "innovators" from the huddled masses of ordinary humans.

One of the absolute best pieces to read on how innovation happens is J. Rogers Hollingsworth's chapter on "High Cognitive Complexity." Synthesis, not separation, is the key.

Update: Kedrosky has written a wonderfully scathing post focused on the CRE part of this issue.

Breaking Through Barriers to Innovation

The search is on for ideas that will stimulate short- and long-term economic growth in the United States without further adding to the budget deficit. Here's a no-brainer: allow university researchers with innovations that are ready to be commercialized to shop around to different technology licensing offices. 

That's the idea advanced by Kauffman Foundation vice-presidents Lesa Mitchell and Bob Litan in the forthcoming issue of the newly revamped Harvard Business Review. This issue features the top "Breakthrough Ideas for 2010" and, happily, this idea comes in at number 6!

Currently, if you are a professor with a new technology or innovation, you typically must work through your own university's technology licensing office (TLO). This is inefficient for a variety of reasons. For one thing, not all TLOs are created equal: some have greater skill than others, while a handful excel. Several decades ago, federal funding of academic research was dramatically skewed toward those universities that engaged in the most commercialization: patents, licenses, spinoffs, etc. In the last twenty or thirty years, the distribution line of federal funding has flattened considerably, with a greater number of schools receiving money than before. This is not to say that the distribution is equal, but it's much less skewed than before.

Here's the kicker: the distribution of innovation output--patents, licenses, startups--has not changed. It's still skewed toward a handful of schools despite the huge increase in the number of universities receiving federal R&D money. We are not reaping the benefits of innovations that are presumably emerging from all that research and development. 

While some TLOs are more competent than others, moreover, it's also likely that such competence varies by area: one TLO may excel at commercializing life science discoveries while another may have a comparative advantage in computer science. Opening up the market for discoveries and commercialization should help iron out the inefficiencies and barriers that exist.

Kudos to Lesa and Bob for a fantastic article and for advancing a top Breakthrough Idea of 2010.

The Not-So-Simple Role of Energy in the Economy

Last week Tim wrote about a New York Times story on "biophysical economics," a small but apparently growing group of scholars who seek to make economics more in tune with the laws of thermodynamics and energy use. Several blogs linked to the story, which I take to mean that this line of thinking was somewhat new to the economic blogosphere. Many of the ideas put forward by these thinkers, however, are not new and if you happen to be interested in things like "net energy" and "energy return on investment" (EROI) and the continuing viability of civilization, you would not have been too surprised by the story.

Tim used the story as the latest installment in his "Pernicious Dystopian Myth" series, which I wholeheartedly applaud and periodically endorse. But for the sake of intra-blog dialogue, I was a bit confused by some of Tim's points. To recap: Tim says "efficiency in energy consumption" will continue," "efficiency in goods consumption is a dominant feature of economic growth," and that since services are less physical than goods, we should expect a less energy-intensive economy.

Let's take these in reverse order. First, services have never been a smaller share of the United States economy than manufacturing. This goes all the way back to 1789. Services have always been an important part of the economy, so it's not necessarily true that more and more services (there has to be a ceiling on this, one would think) will lead to a less energy-intensive economy. Furthermore, services are not necessarily more friendly toward energy efficiency. One of the great ironies of the "intangible" new economy is that many of things techno-utopians like to laud as transformational (the Internet, Web 2.0 companies, software startups, etc) are fundamentally dependent on something as physical as energy. Specifically, electricity and thus coal and water and nuclear power. The computing cloud that is a favorite futurist topic would not exist but for ways to generate electricity. Thus, we have no choice but to remain focused on energy use and extraction and efficiency, notwithstanding our so-called intangible services economy. 

(As an aside, something as cutting edge as IBM's push into services science recognizes the enduring physicality of many service systems.)

Second, yes, efficiency in goods consumption has been a mark of economic growth. This doesn't mean it will continue to be. More deeply, some research does show that we have had to use more and more energy (as well as debt, a subject for another day) to generate incremental increases in economic output.

Tim's third assertion was that energy efficiency in general will continue. Efficiency here is a slippery term; the biophysical economics crowd, as noted in the NYT story, focuses on something called the energy return on investment (EROI), which basically measures the cost-benefit ratio for energy production and consumption. What is the marginal energy return on increasing investments in energy generation? You could still see positive efficiency even if the EROI is narrowing. And, in fact, mounting research does indicate that, at least when it comes to petroleum, the EROI has been falling. We are expending more and more energy to extract energy: the marginal return is falling. So even if "efficiency" appears to continue, it doesn't mean everything is fine and dandy.

In fact, there is nothing dystopian about saying that, pending a negative EROI on the source of energy that makes the world economy run, we are in for a rude awakening when it comes time to transition to whatever comes next. (Natural gas?) It might actually help if the biophysical economists made some headway and expanded the purview of economic analysis to include the things that actually make the economy go. (In this vein, Steven Johnson's latest book, The Invention of Air, includes some delightfully pithy observations about the sometimes invisible, sometimes ignored, connections between what is in the ground and what occurs in the human economy.)

So anyway, Tim's right that pessimism can go too far. But that doesn't mean pessimism is unwarranted.

Pernicious Dystopian Myth, October edition

Try to fill in the blanks (I know, I know, this kind of boilerplate crosses my eyes every few weeks as well, so it could truly be anything ... but this is from a very legit news organization, seriously).

They hope that a set of theories they call "________ economics" will improve upon neoclassical theory, or even replace it altogether.  But even this nascent field finds itself divided, as evidenced by the vigorous and candid back-and-forth debate last week over where to go next. One camp says its models prove the world is headed toward a dramatic economic collapse as ________ takes hold, while another camp believes there is still time to turn the ship around. Still, all ________ economists see only very bleak prospects for the future of modern civilization, putting a whole new spin on the phrase "the dismal science."

Your choices are:

A. Singularity / technological acceleration and automation
B. Rawlsian / globalized labor arbitrage
C. Heterogenous / insolvent overconsumption  
D. Biophysical / energy scarcity

Skip below to the see the answer ...

 

 

 

 

The answer is D.  And it comes from the New York Times, hat tip Free Exchange / Tyler Cowen. And here's more:

In 1926, Frederick Soddy, a chemist who was awarded the Nobel Prize just a few weeks before, published "Wealth, Virtual Wealth and Debt," one of the first books to argue that energy should lie at the heart of economics and not supply-demand curves.

Soddy also criticized traditional monetary policy theories for seemingly ignoring the fact that "real wealth" is derived from using energy to transform physical objects, and that these physical objects are inescapably subject to the laws of entropy, or inevitable decline and disintegration.

The sharpest difference between biophysical economics and the more widely held "Chicago School" approach is that biophysical economists readily accept the peak oil hypothesis: that society is fast approaching the point where global oil production will peak and then steadily decline.

Let's take this seriously, and take it apart carefully.

1. I suspect a great many economists agree with the notion of peak oil, and it is a question of when and how, not if. Given a finite energy source and its increasing usage rate, will it deplete? Yes. The more interesting question is what happens to oil prices, and here I think neoclassical supply & demand is a great tool to understand the consequences.

2. Just labeling your straw man as the "Chicago School" is not much of an argument. Please. Stop.

3. The core argument of BPE is that all economics, and hence economic growth, is about physical goods. That all wealth, all consumption, all value comes from physical things and their production.  This takes energy. And since energy is finite, QED, there is a thermodynamic limit. I think there are four problems in that chain of logic. (1) efficiency in energy consumption will continue (see Obama, smart energy grid), (2) efficiency in goods consumption is a dominant trait of economic growth, and impossible to violate (see the fixed quantity of atoms on Earth, endogenous growth theory, Paul Romer), (3) services are non-physical and an expanding component of GDP (see Poetry, Music, Medicine, Art, Athletics ...), (4) the Sun's energy is finite but not in our time horizon (see Sun).

4. Before the hate mail starts rolling in, keep in mind that I am a member of the Pigou Club. I'd love to see the U.S. replace taxes on labor with taxes on energy. It would do the world a world of good. And help economic growth! 

The Macro War

I have refrained from commenting on Paul Krugman's lengthy critique of macroeconomics which was published earlier this month in the NYTimes Magazine.  I am not sure if I have anything useful to add. It is an important essay that is part of the conversation economists will be having for years to come (which I am calling the "Macro War"). It is quintessential Krugman being Krugman: brilliant, funny, nasty, and a puzzling blend of revelation and overstretch. And forgive me for protesting, but if I see the baby co-op story in print again, I'm submitting Krugman to the authorities for unfairly regurgitating the same pop material to pad his pubs.

If you are genuinely interested in the macro war, do not miss an alternative critique recently published by Northwestern 's distinguished professor Robert Gordon. I highlight five sections here: the entire abstract, three clips on growth vis-a-vis macro, and his assessment of Krugman. It is amazing. (Thanks to Greg Mankiw for the pointer).

Is Modern Macro or 1978‐era Macro More Relevant to the Understanding of the Current Economic Crisis?

ABSTRACT
This paper differs from other recent critiques of “modern macro” based on DSGE models. It goes beyond criticizing these models for their assumptions of complete and efficient markets by proposing an alternative macroeconomic paradigm that is more suitable for tracing the links between financial bubbles and the commodity and labor markets of the real economy.
The paper provides a fundamental critique of DSGE and the related core assumptions of modern business cycle macroeconomics. By attempting to combine sticky Calvo‐like prices in a theoretical setting that otherwise assumes that markets clear, DSGE macro becomes tangled in a web of contradictions. Once prices are sticky, markets fail to clear. Once markets fail to clear, workers are not moving back and forth on their voluntary labor supply curves, so the elasticity of such curves is irrelevant. Once markets fail to clear, firms are not sliding back and forth on their labor demand curves, and so it is irrelevant whether the price‐cost markup (i.e., slope of the labor demand curve) is negative or positive.
    The paper resurrects “1978‐era” macroeconomics that combines non‐market‐clearing aggregate demand based on incomplete price adjustment, together with a supply‐side invented in the mid‐1970s that recognizes the co‐existence of flexible auction‐market prices for commodities like oil and sticky prices for the remaining non‐oil economy. As combined in 1978‐era theories, empirical work, and pioneering intermediate macro textbooks, this merger of demand and supply resulted in a well‐articulated dynamic aggregate demand‐supply model that has stood the test of time in explaining both the multiplicity of links between the financial and real economies, as well as why inflation and unemployment can be both negatively and positively correlated.
     Along the way, the paper goes beyond most recent accounts of the worldwide economic crisis by pointing out numerous similarities between the leverage cycles of 1927‐29 and 2003‐06, particularly parallel regulatory failings in both episodes, and it links tightly the empirical lack of realism in the demand and supply sides of modern DSGE models with the empirical reality that has long been built into the 1978‐era paradigm resurrected here.

(Krugman critique mentioned)

The malaise of modern macroeconomics has recently been the subject of strikingly vitriolic accusations. The Economist (p. 70) quotes Paul Krugman’s LSE Lionel Robbins Lecture of 10 June 2009 as stating “most macroeconomics of the past 30 years was spectacularly useless at best, and positively harmful at worst.” More recently Krugman (2009) has amplified this criticism in a long and overly argumentative denunciation of modern macroeconomics. In notably similar language that might have inspired Krugman, Buiter (2009) wrote three months earlier that “the typical graduate macroeconomics. . . training received at Anglo‐American universities during the past 30 years or so may have set back by decades serious investigations of aggregate economic behaviour and economic policy‐relevant understanding. It was a privately and socially costly waste of time and other resources.”

(Three sections that mention growth are below)

The adjective “business cycle” in front of the phrase “modern macro” represents an important qualification to the scope of this paper. Much of modern macro escapes its critical overview, including such broad areas of modern macro progress as growth theory, search and matching models of unemployment, theories of why prices and wages are only partial flexible, and attempts to use modern versions of production functions and factor input measurement to develop empirical counterparts to longstanding 1978‐era macro constructs such as potential GDP growth and the GDP gap.
...
Viewed more broadly, a more basic similarity between the 1920s and the period between 1995 and 2006 was the view that permanent prosperity had arrived, and that the good times should be allowed to roll. The underpinning of this benign environment was the parallel upsurge in productivity associated with the invention of new “general purpose technologies” (GPTs, Bresnahan and Trajtenberg, 1995). In the 1920s productivity was boosted by the delayed payoff of the invention in the late 19th century of two revolutionary GPTs, electricity and the internal combustion engine. Many of the reasons for the delay in the application of electricity to manufacturing in the analysis of David (1990) apply as well to the internal combustion engine which required technical development and highway construction fully to transform transportation and distribution. Likewise, in the late 1990s the invention of the internet together with other delayed applications of the computer, including bar‐code scanning, boosted productivity growth for several years after the collapse of the late 1990s investment boom itself. More rapid productivity growth held down inflation and in this decade allowed the Fed to justify its maintenance of low interest rates.
...
The required condition for continued full employment is the opening of a gap between the growth rate of nominal GDP and the growth rate of the nominal wage (Gordon, 1984, p. 40) to make room for the increased nominal spending on oil. If nominal wages are flexible, one option is for the growth rate of wages to become negative, allowing the growth rate of nominal GDP to remain fixed. At the alternative extreme with rigid wages, to avoid a decline in nonenergy output an accommodating monetary policy must boost nominal GDP growth by the amount needed to “pay for” the extra spending on oil, but this will lead to an inflationary spiral if expectations respond to the observed increase in the inflation rate. A third alternative, and the one that actually occurred in the 1970s, was a combination of wage rigidity with a partial response of nominal GDP growth, pushing down both real non‐energy spending and employment.

Bill Easterly on Specialization and the Aid Industry

Guest post by Charles C. Johnson

Unveiled in 2000 by the United Nations, the Millennium Development goals have consistently unmet despite pledges from much of the industrialized world. 

On his blog, Bill Easterly of NYU explains why, "Each aid organization tries to meet all MDGs and each fails to specialize. Therefore some aid agencies are forced to supply things they are bad at ... for which there is no demand."

Easterly explains in further detail:

UNICEF is working on swine flu, the traditional province of WHO, who is distracted by trying to do development research, which is the traditional specialty of the World Bank, who is in turn distracted by a new emphasis on children, which is the strength of – just to complete the circle – UNICEF.

Even very small aid agencies fail to specialize – Luxembourg’s $141 million aid budget was divided among 30 different sectors (out of a possible 37). The tiny Luxembourg budget also went to 87 different countries.

With high overhead costs for each separate activity for each country, the ratio of overhead costs to funds for the activity gets extremely high, sometimes over 100 percent. UNDP has one of the very LEAST specialized aid budgets by country and by sector, and it actually does have a ratio of overhead costs to aid disbursed of 129%.

One suspects overhead costs devoured even more of program costs for the $20,000 Greece spent on worldwide post-secondary education, the $30,000 the Netherlands spent on promoting worldwide tourism to developing countries, the $5,000 Denmark spent on worldwide emergency food aid, or the $30,000 Luxembourg spent on conflict, peace, and security. (Remember, these small sums may have been split even further among country recipients.)

One could think of many political economy reasons why aid agencies resist specialization. From my casual experience in a large bureaucracy (the World Bank), the primeval bureaucratic instinct is to give a tiny piece of the pie to every possible lobby group (internal or external). But what’s most clear is that it shows aid agencies lack of accountability, because it is such a wasteful practice that also drives the aid recipients crazy with duplication of efforts by every aid agency in every sector in every country.

I suspect that this problem is even more pronounced than Easterly recognizes. Part of the reason groups specialize, in accordance with Adam Smith, is that they receive a market signal such as a price rise to specialize and thereby innovate, but what evidence is there that there is any such signal in aid agencies? Remember, nobody loses his job for failing to stop poverty.