In the Fall 2015 issue of Cato Journal, Alberto Mingardi (Director General of the Bruno Leoni Institute, and blogger at EconLog) published an article titled 'A Critique of Mazzucato's Entrepreneurial State'. Here I will provide a summary of it.

Just the introduction is a huge condemnation of the book:

Though Mazzucato claims she is building on existing evidence of the effectiveness of government research and development spending, in actual fact her evidence is shaky. She adopts a very extensive definition of industrial policy that includes the unintended consequences of government intervention, and focuses only on 20th century America in making her case for what she deems to be a general law. Moreover, she ultimately fails to prove that the specific government interventions that she hails as beneficial were purposefully directed to achieve the particular outcome in question. This article shows why Mazzucato’s claims for the necessary role of government in promoting an innovation-oriented economy are unconvincing. The fundamental problem is that her work is based on a peculiar line of economic thinking that does not consider the inevitability of tradeoffs while dealing with scarce resources, and does not acknowledge the role of demand and of consumers in a modern market economy.

In 'The Idea of the Entrepreneurial State' section, he points out something already talked about in this blog: the importance of stories, and how Mazzucato's fits some modern preconceptions like a glove, and has garnered praise from basically everyone, thanks not only to her prose, but also to her effective public speaking skills.

For Mingardi, what has cemented her reputation is the debunking of the idea that innovation comes from market interactions, defending instead that the great breakthroughs have been done by the government, not startups or venture capitalists. The iPhone is a product of industrial policy, not free market ingenuity.

Continuing with his focus on vision, he discusses the choice of the US for a sizable part of her analysis: the US is seen as a free market champion, and if even the US rely strongly on the State for innovation, which country won't? She doesn't, however, discuss other examples of not that successful industrial policy in Europe.

Mingardi takes issue with the claim that every (or almost every) innovation owes its existence to the government, declaring it absurd, and its argumentation, self-contradictory.

Then, in 'the Myth of the Entrepreneurial State as an Answer to Austerity', he frames the book in its context: the debate over the role of government in society, and in particular, against 'free market fundamentalists' who say the State should just correct market failures. She wants the State to do more, and so she tries to put forward 'an exciting vision of the State's role', reasoning that if there is a positive attitude towards the State, then the public will not support reductions in its role, or that they will favour expansions of it.

The aim of the book is to 'convince the UK government to change strategy: to not cut State programs in the name of making the economy ‘more competitive’ and ‘more entrepreneurial,’ but to reimagine what the State can and must do to ensure a sustainable post-crisis recovery'. She attacks the market failure-fixing approach (and the Public Choice School) as too simplistic. But that critique is nonsense, says Mingardi: PCS occupies itself with describing how policymaking works, not innovation!

What would be necessary for her to be right?

In order to succeed in her own argument, Mazzucato needs to prove two things: first, that there is a treasure trove of examples showing that government intervention is ubiquitous in the history of modern capitalism (a point that won’t be controversial); and second, that a particular kind of government intervention—industrial policy—has, consistently with its own declared goals, been effective in fostering innovation. Is the historical evidence that Mazzucato provides robust? Or is it just an ex post rationalization of a patchwork of policies that were not necessarily put in place to promote a particular research program?

In the next section, 'Industrial policy', he explicits a vital component of Mazzucato's argumentation: she's not just saying governments help. She's saying they are necessary. She tries to argue that this is so on basis of certain historical experiences, and then she argues that it ought to be so, as the private sector wouldn't be able to do what is necessary.

That is, Mazzucato analyses a period of massive expansion of governmental activity in the economy, and it would be improbable that some of that spending wouldn't have lead to some innovation-producing businesses.

However, this wasn't the case in the 19th century: there was little government investment and plenty of innovative activity, so she cannot generalise her argument just like that.

Furthermore, she needs more than government investments conducive to innovation to prove her points,

Mazzucato should persuasively show that the innovations she claims are due to government intervention are the result of “intelligent design.” It is clear that she believes this to be the case, as she argues against the opposite view: “We are constantly told that the State should have a limited role in the economy due to its inability to ‘pick winners,’ whether the ‘winners’ are new technologies, economic sectors or specific firms” (Mazzucato 2013: 18). In her perspective, however, the role government should exercise is precisely “directionality (choosing areas of change, rather than just ‘facilitating’ it)” (Mazzucato 2014: 4) in R&D investments. Her narrative is one of “a confident state that was able and willing to courageously envision the direction of change-defining missions and to organise institutional structures across public agencies and departments” (Mazzucato 2014: 7).

Her book offers many examples of the role public policies played in promoting innovation. And yet, these innovations may often be considered as positive externalities of public intervention, as opposed to carefully designed outcomes of such industrial policies. This is problematic if you are eager to defend a government that picks winners: Surely you should first demonstrate that it did actually pick winners, because this is the “mission-oriented directionality” on which Mazzucato’s case hinges.

Then, in 'Did Government Invent the Internet?', Mingardi points at some examples she uses to defend the idea of an Entrepreneurial State: DARPA, the SBIR program, orphan drugs regulation, and nanotechnology. With these cases, she argues, the State did all of the hard work, and then the private sector picked the low-hanging fruit and reaped all of the rewards.

But Mazzucato skips details, and we need those to assess the relevance of those institutions or programmes. As an example, Mingardi explains the origins of the Internet.

The popular story is that 'the US government did it'. More educated people may go further and say that 'DARPA did it'. But not quite. While it is true that the federal government supported work in what would become the building blocks of the internet (like FTP, or TCP/IP), some of those who developed key concepts behind the internet, Licklider and Kleinrock did before joining DARPA. It was after than when they were hired by DARPA, so it wasn't that the government asked for packet switching, but that some people were working on that, and DARPA decided to go for it. Furthermore, the internet is not just algorithms: the physical hardware, such as routers or optical fibers were developed by private enterprises such as Cisco or Corning Glass Works.

Regarding the grants that funded work leading towards the internet, Mingardi also has things to say:

Historian Price Fishback (2007: 516) admits that“no one can deny the vast repercussions of militarily motivated activities” in the development of the Internet. He goes on to say, “The military’s role was clearly sufficient to develop the early technologies, but arguably it was not necessary. The credit for these technologies should go to the actual people performing the research” (521).

The relevant question is thus whether the development of the Internet took place as the result of some “mission-oriented directionality” on the part of government, or if it is better seen as merely a positive externality of public intervention.

According to Fishback (2007: 519), “The military funding contributed spillover benefits to the development of the commercial Internet” by not trying to tightly control the projects, by encouraging wide dissemination of research results, and by funding small firms. This suggests there was little “mission-oriented directionality” behind the creation of the Internet

Also, universities in the US, unlike those in Europe, were trying hard to understand societal problems and research things that were relevant for that:

Like Mazzucato, Nathan Rosenberg (2000) points out the important role played by American universities in developing prototypes, and more generally in basic research. However, unlike Mazzucato, he acknowledges that they are very responsive to the needs of the economy and of society at large, showing greater flexibility than their European counterparts.

The “competitive environment” in which U.S. universities operate (Rosenberg 2000: 38) could be an explanation for their high research productivity. It certainly is relevant to note that government funding is channeled through institutions that compete with one another to attract capital (either public grants or private donations), as well as to attract the best factors of production (teachers), and also to gain customers (students).

Here, as so often in her book, Mazzucato simply assumes that if something goes right, government must be responsible. But in the real world the mere existence of government money doesn’t account for the different nuances of institutions. Government money channeled through a competitive setting may have very different effects than government money spent following a strictly hierarchical, topdown logic.

Next section is 'Can Industrial Policy be decentralized?'

Mazzucato talks about a decentralized form of industrial policy when discussing DARPA, but according to Mingardi, this cannot be: Industrial policy is about the government deliberately promoting industry. There is a difference between mission-oriented innovation, and innovation that came to be as an unintended consequence of some particular programme.

With the SBIR programme, there doesn't seem to be government guidance in there: it's a requirement to large federal agencies to spend money on small businesses. Mazzucato argues that the SBIR programme is now larger than 20 years ago because of the retreat of venture capital -and SBIR had to step in to cover the gap- but Mingardi counters that just because some public programme grows doesn't mean it's succeeding in fulfulling its role. He cites Tanzi 2012, saying that government programmes in general tend to grow and become more expensive. Thus, it is possible that SBIR has not grown because it was a necessity for a stable rate of innovation, but just because of the same dynamic that applies to any government programme.

He then goes for the iPhone example. He argues that Mazzucato sees industrial policy even in the most trivial things. For example, in the case of the History of touchscreens, the government funded Wayne Westerman's PhD education, and then he went on to cofound an important touchscreen company. Are we to believe that was an example of industrial policy? Of government-led innovation?

In the next section, 'What about Entrepreneurs?', he says Mazzucato attempts to answer the question of where innovation comes from. From a market-nourished intrinsic propensity for entrepreneurial discovery and innovation, is Mingardi's answer, channeling McCloskey and Kirzner. On the other hand, Mazzucato's answer is government investment in R&D. She says that, for example, Japan succeeded where the Soviet Union failed in generating prosperity not because they spent more in R&D, but because Japan had strong user-producer linkages. Meaning, they had a market economy. But she is attributing Japan's success to industrial policy while she could be also attributing it to the presence of a free market, yet she doesn't do that. If History has ever seen an Entrepreneurial State, surely that was the Soviet Union. Everything the soviets did was done within the framework of the State, and yet they weren't able to compete with the US. The Soviet Union also serves as a rebuttal to the often held position that military spending is a way to prosperity. It could be that it is, coupled with something else, but arguments for that have to be provided.

Japan, says Mingardi, had a consumer oriented industry, and until 1991, less than 20% of its R&D spending was governmental.

She also, in emphasising breakthrough innovation, absurdly downplays the role of private enterprises like Apple, making they seem like mere repackagers of State-funded science. It is this view that fuels her proposals of making Apple pay more taxes: if it is a parasite that has done little but to surf the governmental wave of innovation, shouldn't some of the rewards accrue to the government instead of to Apple? (No).

Furthermore, there is a difference between inventing things, and creating consumer-ready products:

What private business does, in a market economy, is order factors of production in a way consistent with its attempt to meet and anticipate consumer demand. Breakthrough innovation doesn’t happen in a vacuum and is seldom realised just because of brilliant ideas and new technological achievements. “Gadgets” alone are not the be-all and end-all of innovation. To be successful, they must also create excitement among buying customers, meet a demand, and thus cause a readjustment of the factors of production. Technological progress doesn’t add new products to the shelves by itself.

Mazzucato sometimes points out that the supposed innovation stagnation we are living in is caused by a lack of interest from big companies in basic science (so the government has to step in). But there are opposite arguments: Maybe it's not that there is less innovation, but that it is occurring elsewhere (the world of computer science) because there is too much regulation in the 'world of things'.

The Entrepreneurial State, says Mingardi, is no such thing. Entrepreneurship is about discovering needs, consumer preferences and better ways of satisfying them. Mazzucato's State is not subject to the same feedback mechanisms as the real entrepreneurs.

This theme is expanded on in the next section, 'The Missing Consumer', in which Mingardi notes that Mazzucato fails to address counterarguments, because

In her view, “We just haven’t developed the accurate metrics needed to judge its [the State’s] investments fairly” (Mazzucato 2013: 19). Moreover,

<<What is ignored is that, in many of the cases that the State “failed,” it was trying to do something much more difficult than what many private businesses do: either trying to extend the period of glory of a mature industry (the Concorde experiment or the American Supersonic Transport project), or actively trying to launch a new technology sector (the Internet, or the IT revolution)>> [Mazzucato 2013: 18].

With Solyndra, she says the problem was with private investors, they didn't back the company when it was having hardships. Instead, the US government should have acted like

the Chinese government, which nationalized Suntech’s assets, aiming “to protect the interest of thousands of workers, the public banks backing the firm, and the State” (Mazzucato 2013: 154). Here, finally, Mazzucato sides with something we can unambiguously recognize as industrial policy. Such an industrial policy implies a defense of the way factors of production are employed today, thereby picturing it is the best possible way, against traumatic reallocation due to bankruptcy.

For Mazzucato the key is that the government need not care about returns on investments because it is funding itself with taxes. But if government redirects resources towards one place, those won't be available elsewhere. Yes, private firms are not infinitely long-term oriented, but why should they be? There are tradeoffs there, but with a big difference: if a company takes a mistaken decision, they receive a strong signal: market losses and ultimately, bankrupcy. If the same happens with the government, not quite, as taxes do not depend on governmental performance.

Mazzucato proposes a set of policies to enable the government to earn some of the rewards that are, according to her, unjustly accruing to the private sector: direct government investment, a bigger role for government banks, and a "golden share" over patents that resulted from public funding. But that will need funding, and she is not analysing the effects of taxation on companies. She defends her policies on the assumption that the lack of an Entrepreneurial State will be harmful to the progress of innovation, but she doesn't even try to quantify the social costs of that.

In conclusion,

Mazzucato doesn’t really explain how government bureaucracy can lead innovation with “mission-oriented directionality.” Furthermore, she doesn’t appear to see innovation as anything other than technological progress per se. The fact that innovations should become “products,” available to benefit flesh-and-blood human beings, doesn’t seem to be particularly relevant to her argument.

Of course, if we assume that we can pay for all possible research projects, then we can safely assume too that financing everything will lead, at some point, to some results. But the U.S. Treasury does not work like Mary Poppins’ bag. So how does this all fit into a world of scarce resources and inevitable tradeoffs?

In such a world, the dreary judgments of investors help at least in figuring out which kind of technological advancements promise to be of use to consumers, and which do not. Consumers are not mere passive subjects: their preferences and needs influence production too. In this respect, Mazzucato’s construction of the almighty entrepreneurial state seems to miss something fundamental. Her lack of consideration of the role consumers play in a market economy points to a conception of economic life similar to the race between the United States and the Soviet Union to send the first man into space. Like those cold warriors before her, Mazzucato overlooks the fact that innovation is not just about technological progress for its own sake, but rather about making people’s lives better and easier. In the end, Mazzucato’s entrepreneurial state, for all its progressive zeal, seems ill-suited to that important task.

Comments from WordPress

  • Philo Fido Philo Fido 2016-11-14T01:39:50Z

    It is written: "However, this wasn’t the case in the 19th century: there was little government investment and plenty of innovative activity, so she cannot generalise her argument just like that." This is ignoring the fact that 19th century economies were not *inclusive growths". Mazzucato makes the case that state funded breakthroughs should also socialize rewards...