Sunday, September 18, 2011

State Intervention and the Computer

There is a perennial debate about the history of the computer, and the extent to private enterprise or government intervention was the driving force behind its development. Of course, both were involved, but the influence of government was very important and most probably the major factor:
“Transistors were developed in a private laboratory, Bell Labs, but it was a monopoly. There was no market and there was no consumer choice. Since it was a government-supported monopoly, they could charge monopoly prices, which is in effect a tax. As long as they had the monopoly, Bell Labs was a very good lab. They did all sorts of things at public expense. As soon as it was deregulated, Bell Labs went down the tubes.

Quite apart from that, Bell Labs was using state-generated wartime technology. Furthermore, they had nobody to sell the transistors to. For about ten years, the only market for high-quality transistors was the government, just like computers. During that period they were able to develop the technology, the scale, the marketing skills so that finally they could break into the market system.” (Barsamian and Chomsky 2001: 18).
In addition, the development of modern computers also occurred in Japan, where this happened by direct Japanese government industrial policy through their Ministry of International Trade and Industry (MITI):
“The real success of Japanese producers, American industry sources conclude, came only after the mid-1970s. [The Ministry of International Trade and Industry] targeted the computer and telecommunications markets as central to Japan’s future. Establishing a national goal to lead in those industries, the government offered substantial incentives to encourage R&D and investment, besides restricting foreign access to Japanese markets.” (Okimoto et al. 1984: 17).
The Japanese industrial policy was so successful that America semiconductor manufacturers were under extreme pressure by Japanese competitors by the 1980s (Nester 1997: 115–116), especially by the dumping of dynamic random access memory chips (DRAMs) on world markets. The American response to that Japanese threat was protectionism in the form of the Semiconductor Trade Agreement (STA) in July 1986 and tariffs on Japanese electronics imports in April 1986, to give US manufacturers relief. This eventually allowed them time to innovate and develop new generations of microprocessors and regain market share.


Barsamian, D. and Chomsky, N. 2001. Propaganda and the Public Mind: Conversations with Noam Chomsky, Pluto Press, London.

Nester, W. R. 1997. American Industrial Policy: Free or Managed Markets?, Macmillan, Basingstoke.

Okimoto, D. I. et al. (eds). 1984. Competitive Edge: The Semiconductor Industry in the U.S. and Japan, Stanford University Press, Stanford.


  1. Do you know if there's a meaningful definition of "private company", where companies simply existing in a country and making use of its infrastructure might be "private", but a company which has the government as its most significant or only customer (Bell Labs, Harris, etc.) would be considered "public" or "nationalized"?

  2. I suppose it would be a private company that is profitable only because of government purchases - one reliant/dependent on the state sector.

  3. Yes. As the debate you had at that Austrian blog shows, though, you can emphasize certain inventions over others to the point where you just go around in circles. (It should be noted your opponent chose not to cite computer science textbooks, or engineering textbooks, or anything at all, and cited himself continuously.) It is clear the state certainly had a role, though, and probably the major role in certain generations.

    I bring it up because it throws a monkey wrench into the Libertarian belief system in the following ways: (a) it brings into question the whole notion of who is the true owner of property; (b) it shows public systems can accomplish things without market discipline; (c) it shows the public system can often outperform the market system.

    For example, it's pretty absurd to believe that the government doesn't have the right to implement taxes on things it handed over to the private sector in the first place, which the public funded, such as the rights to broadcast digital television. Of course it has a right to tax things or put things in public control that it created.

    Furthermore, it shows the collusion corporations have with governments. This in some ways justifies taxation: when you go to work for a corporation, you're entering into an agreement with the corporation and the government which protects it (and likely helps to fund it). Thus, taxation is as "voluntary" as choosing whom you want to work for, and what you buy, etc., since the system is arranged by the government in the first place.

    I think Chomsky brings it up because of (c), that the public system is often the best way to get things done, which is debatable I guess, but (a) and (b) are certainly not debatable.

    "The law creates and protects that bundle of rights called property or the corporation, and this same law can rearrange that bundle of rights if it is in the public interest..."It hardly seems valid to condemn the government for legally rearranging this bundle of rights when it created them in the first place."

    - Ralph Nader

    Libertarians bring up "big government" all the time, and they claim that this "big government" probably had no role in shaping industry and that they would have done it that way on their own. Again, the message is completely incoherent.


  4. Here is some more stuff on public funded research. Of course it is not just in computers -- the entire biotechnology sector got its start with public funding, and, a great deal of technological progress has been underwritten by the government (Chomsky gives the figure of about 85% in Understanding Power).

    Chomnsky likes to say this has effects on the entire market. For example, say you have some big corporate district that exists because of land giveaways, land grants, redistricting, and zoning laws, and so on. Where is the guy with the hot dog stand (probably one of the few true free-market areas in the world) going to go? He's going to go to the corporate district because that's where people are. So this intervention trickles all the way down.

    From Nader:

    "The U.S. federal government is quite probably the richest property owner on earth. The government owns vast tracts of land, including oil and mineral riches, forests, thousands of buildings and plants, the public airwaves and much more.
    Because they often do not appear as budgetary debit items, government giveaways too frequently escape the corporate welfare stigma. Giveaways are in fact one of the purest forms of corporate welfare -- a something-for-nothing, or something-for-too-little, proposition. The level of public outrage would be high if the government wrote a $70 billion check to the broadcast industry -- but that is effectively what happened when the Federal Communications Commission, pursuant to the Telecommunications Act of 1996, handed over the digital television spectrum to existing broadcasters.
    As former Senate Majority Leader Bob Dole has recognized, there is no conceivable reason why the incumbent broadcasters should have been given exclusive rights to use the airwaves.7 Other possible television broadcasters should have been given the right to bid for portions of the digital spectrum, and so should have other potential users, such as data transmission companies. "
    "In one of the single biggest giveaways in U.S. corporate welfare history, the Federal Communications Commission (FCC) on April 7, 1997 donated broadcast licenses for digital television to existing broadcasters.
    Under the terms of the giveaway, the broadcasters will pay nothing for the exclusive right to use the public airwaves, even though the FCC itself estimated the value of the digital licenses to be worth $11 billion to $70 billion.6
    The giveaway was mandated, in part, by the 1996 Telecommunications Act, which prohibited, under demands by the broadcaster lobby, the FCC from auctioning off the airwaves. The Telecommunications Act also required the FCC, if it decided to allocate the licenses, to give them only to incumbent broadcasters. "


  5. In 1996, Congress quietly handed over to existing broadcasters the rights to broadcast digital television on the public airwaves-a conveyance worth $70 billion-in exchange for... nothing.

    Although the public owns the airwaves, the broadcasters have never paid for the rights to use them. New digital technologies now make possible the broadcast of digital television programming, (the equivalent of the switch from analog records to digitized compact disks), and the broadcasters sought rights to new portions of the airwaves. In recent years, the Federal Communications Commission has, properly, begun to recognize the large monetary value of the licenses it conveys to use the public airwaves-including for cell phones, beepers, and similar uses-and typically auctions licenses. The 1996 Telecommunications Act, however, prohibited such an auction for distribution of digital television licenses, the most valuable of public airwave properties, and mandated that they be given to existing broadcasters.

    "The federal goverment has historically funded about half of all U.S. expenditures on research and development (R&D) -- some $74 billion in fiscal year 1992. Over the past twelve years, the allocation of property rights in these research projects has dramatically changed. Before Reagan/Bush, the government genreally sought to have research products enter the public domain, or to patent its inventions or license them on a non-exclusive basis. Exclusive licenses were used, but only sparingly, and often for limited terms. After 1980, however, a series of statutes, rules, and policy memoranda sanctioned a broad use of exclusive licenses. In effect, taxpayers invest billions of dollars in R&D every year -- and then the returns on these investments are privatized." --Ralph Nader, Nader Reader, page 26.

    The megacorporate world of Ronald Reagan is a good piece.


  6. I suppose it would be a private company that is profitable only because of government purchases - one reliant/dependent on the state sector.

    Well, my point was that such a definition of "private" is disingenuous. I wouldn't go as far as excluding a company from "private" status if it uses public resources, because every company does. But calling companies like General Atomics, Lockheed, or Raytheon, or universities like Harvard "private", even with caveats, just isn't honest. They're not private in any meaningful sense: they could not continue to operate normally if they lost federal funding. They're also too autonomous to be considered "public", or even a "government subsidiary". "Defense contractor" covers some cases, but doesn't really capture the idea. I don't know what to call them.

  7. I don't understand this idea of a policy autarky - with X policy there is success, without X policy there is failure.

    Plenty of protected industries exist in the Third World under the Sick Industrial category in sub-Saharan Africa, India, West Asia, and whatnot, which are still failing to sell or produce.

    Many textile businesses are classified as sick industries in India, and are given every benefit of industrial policy possible for six decades. They are still struggling, loss-making industries.

    We simply have to remember that

    a) it is the customer who is the ultimate determiner of success of an enterprise
    b) a customer's choice could even be a matter of an arbitrary whim and may tend in any direction
    c) even if government sets policy, the enterprise has to sell to customers, not the government, and policy does not automatically translate to sales

    Whether or not there would have been an industrial policy for semiconductor businesses, the ball was always in their court to find and keep customers. Industrial policy seems merely a matter of push-and-pull.

    Either way, it's not like an American consumer cares whether his computer has a Japanese or American semiconductor, and his welfare is not increased or decreased by nationality of product. Industrial policy is based on the arbitrary idea of business having a nationality, as if a poor quality product from a home country is a boost for my welfare.

  8. "the development of modern computers also occurred in Japan". This is not true. The Japanese flooded the DRAM (computer memory) market. Mini and Micro computers were invented inthe US. The Microprocessor was invented in California by Intel and this component was the key for the independent and garage based creation of the PC by the homebrew club and later Apple (Jobs and Woz). There has never been a successful Japanese microprocessor no matter how hard they tried. By the time the US santioned DRAM imports here were no longer made in the US at all (and never have been again). It was too little and too late. Read Andy Groves book Only The Paranoid Survive to learn the history. Then the Japanese ended up losing the memory market to Taiwan and have never amounted to much in computing. Also their state corporate tehlchnology economy there has flatlined for over two decades so what's the point in highlighting it?

  9. "Also their state corporate tehlchnology economy there has flatlined for over two decades so what's the point in highlighting it"

    If you're referring to the "lost decade", that didn't extend for "over two decades". That is utter nonsense. It ended by about 2004.

    And its cause was a bursting asset bubble and debt deflation. It had nothing to do with Japan's industrial policy.

  10. By the time the US santioned DRAM imports here were no longer made in the US at all

    That is also false.

    In 1990, Motorola, Micron Technologies, and Texas Instruments were manufacturing DRAMs.

    (National Research Council, Keeping the U.S. computer industry competitive, p. 21.

  11. Prateek Sanjay,

    The fallings of India's industrial policy are well known.

    You can read a plausible explanation of why it was not succesful in the way that, say, South Korea was in Vivek Chibber, Locked in Place: State-Building and Late Industrialization in India, Princeton University Press, Princeton, N.J. 2003.

    There are instances when modern scientifically-based medicine does not treat patients even when the treatments in question work in many other cases. The proper response to that is to ask why it didn't work in individual cases, not overthrow scientifically-based medicine and replace it with a lassisez faire (do nothing) approsch to disease treatment.

    There are so many empirical instances of the success of industrial polciy or infant industry protectionism, individual cases of failure warrant careful examination. I think Chibber explains why it did so in India's case.

  12. This is drifting away from the original discussion.

    Again, there is no such thing as X industrial policy -> Success, Y industrial policy -> Failure.

    Because the customer is the lone determinant of the success or failure of industry. No amount of industrial policy will protect a business that has no customers. No amount of industrial policy will boost a business that is unable to understand large buyer's or customer's needs and market to them accordingly.

    You as a radical subjectivist should more likely have taken this position, so it's only more baffling.

    The point is that it is a waste of time to discuss industrial policy as a motivator, because a) it's not the government who is the buyer here, b) it's not the government who is the producer here, and c) hence, government is one minor player that has the least influence in whether something gets produced and sold.

  13. "c) hence, government is one minor player that has the least influence in whether something gets produced and sold. "

    That is not true.
    The government sector and it purchases are a major sector of the economy in most countries.

  14. Okay, what I meant to say is that industrial policy is the least influential on whether something gets produced or sold. Industrial policy will not somehow turn underproductive methods of production to produce more or somehow create demand for goods that nobody wants to buy.

    A simple illustration is the fact that General Motors enjoys an unlevel playing field where it has more benefits in the American markets than Japanese manufacturers do. None of those benefits prevented GM's market share from falling from 50% to 25% or Japanese manufacturers to continually outdo them in their own markets.

    Irrespective of what tarriffs or subsidies exist for GM, if an American purchaser wants to buy a Japanese car, HE WILL BUY A JAPANESE CAR.

    In short, your discussion of the governmental role here is IRRELEVANT, because what determines whether or not a car can sell is:

    1) the buyer's tastes and preferences in cars
    2) the branding of the automobile manufacturer
    3) the safety, speed, and functionality of the car, along with other features

    I believe that many of you industrial policy institutionalists seem to imagine a purely abstract world based on economic ideology only. Whether American manufacturers fail or succeed has somehow nothing to do with management quality, worker productivity, marketing of the products, or quality of products - for you, it is purely a matter of whether enough subsidies were in place and whether enough tarriff walls were erected. In your discussion of business, you discuss everything but business.

  15. 1) the buyer's tastes and preferences in cars
    2) the branding of the automobile manufacturer
    3) the safety, speed, and functionality of the car, along with other features

    You left out a couple of things. Price and availability both of imports and locally manufactured goods are directly influenced by industrial policy.

    Not to mention quality, which is often influenced by using export bounties, which require competition with foreign firms, and also through simply setting minimum standards, such as a requirement that cars can reach 150 miles per hour, or travel 50 miles per gallon.

    During the cold war, the united states had a military directed industrial policy, with local sourcing rules and elaborate programs to develop strategically important technology.

    I wonder if you've read any work by say Ha Joon Chang, or Ian Fletcher

  16. I have read papers by both of them. And Ian Fletcher is a corporate lobbyist, for god's sake.

    Look, I know the whole, "Look at the arguments, not the person" issue.

    But the fact that a leading industrial policy institutionalist is a corporate lobbyist is a reminder that

    a) industrial policy is all about how many benefits a corporation can leach off the system
    b) industrial policy is a way of ensuring that the government picks you as the winner, instead of markets

    Unlike most other economic thought, this doctrine is the most shamelessly and explicitly oriented towards serving narrow agendas.

    PS: A price is merely a reflection of the customer's subjective preferences of a product. Even if you artificially raise prices of foreign products, and lower prices of domestic ones, a customer who prefers the foreign product will still buy it. It again comes back to the fact that industrial policy is completely pointless.

  17. "A price is merely a reflection of the customer's subjective preferences of a product."

    I disagree. Though in markets where you can haggle or competitively bid they that may have some truth, in many markets long-run, normal prices tend to be explained by the cost of production plus the profit markup. While zero or insufficient demand means a business can't make a profit, that still doesn't mean that subjective preference of consumers is the only factor causing price.

    "Even if you artificially raise prices of foreign products, and lower prices of domestic ones, a customer who prefers the foreign product will still buy it."

    Maybe, but what of many consumers who prefer the cheaper one? Do you really belive tariffs have zero impact on consumer decisions?

  18. "Do you really belive tariffs have zero impact on consumer decisions?"

    And I would add; if not, how are domestic producers wringing any benefit out of this arrangement?

    Sure, Ian Fletcher is a corporate lobbyist. There are plenty of free-trade economists sucking at the corporate tit as well, often affiliated with privately funded "think tanks" such as the Hoover Institution, and the Mercatus center, or the Cato Institute. They're getting their payola from a different set of elites. That doesn't make them objective. If you want to turn the discussion into a series of ad-hominems against authors, that's fine with me. I actually enjoy flinging mud more than rational debate, so have at it!

    Back to your example of the Japanese car manufacturers and their superiority to US firms. These manufacturers were a product of one of the most sophisticated industrial policies ever put into practice... and subject to some of the most powerful trades unions.

    Meanwhile, in the US, following world war II, tariffs and industrial policy in the civilian sector were gradually abandoned. In the free traders' view, american car makers should have been the most competetive and efficient producers of the best products, since they were less coddled than the japanese car makers, for example. Somehow, this proved not to be the case.

  19. Businesses do have products sold at a loss in order to keep their market share. Videogame consoles such as Playstation or XBox are often sold at a loss. Easily verified with a quick Google search, btw.

    And technically, a lot of pharmaceutical drugs are sold at a loss, once you exclude the per unit cost of research. Research costs are sunk costs, and businesses might not bother recovering sunk costs.

    Besides, the word "cost of production" is never easily defined. Are fixed manufacturing overheads (which don't vary according to the level of production) such as factory rent or electricity a part of the product cost or a non-production expense? As a student of cost accounting, I have learnt that this is purely subjective.

    So as far as I can see it, it's really just a customer defining prices by choosing to buy or not buy at a particular price level.

  20. Furthermore, if some specific industry is being driven out of business by some foreign company engaged in dumping, what do you do? Allow your industry to just collapse?

  21. Before we even discuss dumping, and whether or not it is good or bad, how do we prove that dumping even exists?

    Supposedly, a dumper is one selling below cost.

    How do we verify that? Travelling to China or wherever the dumper? Visiting a dozen factories there? Determining all the input costs? Determing all the daily output? Apportioning all the fixed costs for each unit of the product? Checking prices?

    Impossible. Prices vary according to day to day, place to place, and time to time. There is no way of ascertaining whether dumping even exists.

    I am going to use search to see if you have anything written on dumping. I wonder about your views on the matter.

  22. "how do we prove that dumping even exists?"

    it's easily googleable;)