Can we model creativity?

April 1, 2009

by Roger Koppl

The issue of creativity has arisen in a fun discussion on Schumpeter over at The Austrian Economists. I have often heard people say that we cannot model creativity. I suppose that must be true in some sense, and yet there is much we can say about “creativity” and social institutions. Early in the Wealth of Nations Adam Smith gives us a good account of what creativity is, namely, the “combining together the powers of the most distant and dissimilar objects.” As my old friend Dick Langlois taught me years ago, creativity is recombination. I think we can have an economic theory of recombination. Indeed, the elements are out there.

William Baumol’s famous 1990 article on entrepreneurship gives an important part of the analysis (“Entrepreneurship: Productive, Unproductive, and Destructive,” JPE, 1990, 98: 893-921). His abstract:

The basic hypothesis is that, while the total supply of entrepreneurs varies among societies, the productive contribution of the society’s entrepreneurial activities varies much more because of their allocation between productive activities such as innovation and largely unproductive activities such as rent seeking or organized crime. This allocation is heavily influenced by the relative payoffs society offers to such activities. This implies that policy can influence the allocation of entrepreneurship more effectively than it can influence its supply. Historical evidence from ancient Rome, early China, and the Middle Ages and Renaissance in Europe is used to investigate the hypotheses.

Institutions create incentives that influence the direction of entrepreneurship. Incentives influence which distant and dissimilar objects are combined.

Baumol seems to hold that entrepreneurial alertness is roughly constant across institutional settings. But both Benny Gilad and David Harper use the idea of “locus of control,” which shows how the institutional structure can influence the overall amount of entrepreneurship. (See Gilad, B. (1982). On encouraging entrepreneurship: an interdisciplinary approach. Journal of Behavioral Economics, 11(1), 132–163 and Harper, D. (1998). Institutional conditions for entrepreneurship. Advances in Austrian Economics, 5, 241– 275.) You have an “internal locus of control” if you think outcomes depend on your actions. You have an “external locus of control” if you thin outcomes depend on outside elements beyond your control. Persons in psychiatric hospitals may easily acquire an external locus of control because others plan everything for you. What you eat, where you sleep, your clothes, the temperature of the room, and so on all depend on choices made by someone else, not you. In such and environment, your entrepreneurial alertness may switch off. In a liberal policy regime of free contract, rule of law, and so on, your outcomes depend on your actions to a relatively high degree. It is a “mixed game of skill and chance” as Hayek says somewhere, but your actions are an important factor influencing outcomes. In such an environment, entrepreneurial alertness is switched on. We get more alertness and creativity. We get more and better combinations of distant and dissimilar objects.

Finally, I think there is also a complexity and computability dimension to the issue. We use the words such as “creative” and “novel” only for unexpected instances of recombination. Although “creativity” is “mere” recombination, we do not see all the creative possibilities at once because we cannot make all the computations required for such omniscience. If we have a relatively simple system, we will have few unexpected instances of recombination and little creativity. If we have a more complex system, more instances of recombination will be surprising. An important result by K. Vela Velupillai shows how this works. In “The impossibility of an effective theory of policy in a complex economy” (in Salzano, M. & Colander D. (eds.), Complexity hints for economic policy, 2007, Berlin: Springer) Velupillai shows that if the economy is “complex” in the sense that it is “a dynamical system capable of computation universality,” then an “effective theory of policy is impossible.” That means that we cannot predict what the system is going to do. If you want to know what it is going to do, you have to watch it unfold. Perhaps the system is “deterministic” in some metaphysical sense. God’s infinite mind is not in the least surprised by what happens. But we mortals with our finite minds cannot make godlike predictions of the system’s behavior. Indeed, it is logically impossible for us to make such predictions if the system is “complex” in the specified sense. Thus, we are constantly surprised by its creativity, by the distant and dissimilar objects the system combines together.

I think policy can influence the amount of creativity by ensuring things such as transparency that enable increased complexity and foster internal locus of control. Policy can influence the direction of creativity for the sort of reasons Baumol explained. Does that mean we can, after all, “model creativity”?

9 Responses to “Can we model creativity?”

  1. Bogdan Enache Says:

    I think you provide the answer to your own question, but I wanna add that the broadly called institutional environment is only one side of the coin. Even if you provide the best lab to some very motivated scientists, for instance, plus 50% share in profits or, for those who value glory more than money, a statute in every town in the country, you still have absolutely no certainty that they will stumble upon the mystery of the universe, as it were.

  2. liberty Says:

    “Persons in psychiatric hospitals may easily acquire an external locus of control because others plan everything for you. What you eat, where you sleep, your clothes, the temperature of the room, and so on all depend on choices made by someone else, not you.”

    This extreme case may be true (although some people in prisons and hospitals get very creative in trying to escape, trying to obtain goods from outside the institution, or in setting up a gang inside, etc.)

    However, the planned economy is the closest economic system to the environment you describe, and I am not sure that the “external locus of control” was created even there. There were very few economic entrepreneurs, but very many political entrepreneurs, climbing the ladder of Party success, through socialist enterprise, through the Party and through the state.


  3. Roger: On an economic theory of recombination, Kirzner tries to do this in Capital and Interest (at least that’s how I interpreted it!)

  4. Peter Klein Says:

    Roger, there’s a lot of work on creativity among (non-economist) management and organizational scholars that might be worth consulting. Check out Keith Sawyer’s blog, for instance:

    http://keithsawyer.wordpress.com/

  5. koppl Says:

    Liberty,

    Institutions affect both direction and magnitude. Thus, entrepreneurship was concentrated on politics in the old Soviet system. But there may have been less of it altogether. According to cliche, folks in old Soviet system were more inclined to wait for orders, exhibit fear of innovation, and so on. A similar claim has been made about people under Nazism. Many of my Italian friends complain that their relatives of old enough to have grown up under Fascism resist innovations of all types, exhibit an exaggerated rigidity of habit and mentality, and so on. I tend to think these cliches and these gripes of my friends capture something real. And I worry that our current rush into deeply politicized markets here in the US will divert American entrepreneurship into the political process. It’s a grim thought, really.

  6. liberty Says:

    Roger-

    Perhaps. But it is hard to measure. Waiting for orders was PART of climbing the political ladder in the Soviet Union. Innovation was not. But, to be a great political entrepreneur, you still had to be creative and have an “internal locus of control.”

    For example, an entrepreneurial manager in the Soviet Union did things like: read Pravda with a very politically wise eye to read between the lines for hints about upcoming priority “campaigns”; the entrepreneurial tolkachi supply expediter had to find scarce inputs, locate pliable managers and determine how to coax them out of the managers (with blat) and build reputations with as many of these factories as he could…a lot of entrepreneurial schmoozing…
    see e.g. (Berliner 1957)

    There was a huge reward to doing these things properly: starting with money (as Tolkachi or successful manager), then scarce goods and prestige at a higher level of office based on successful plan fulfillment, then the really scarce stuff and the power, once in with the party, and finally the pinnacle was to become part of the Nomenklatura…

  7. Daniel Klein Says:

    I agree with Koppl, who starts his post: “I have often heard people say that we cannot model creativity. I suppose that must be true in some sense, and yet there is much we can say about ‘creativity’ and social institutions.”

    What is meant by “a theory of creativity”? If we think theory as explanation, then what is explaining what? Is creativity the explanandum or the explanation?

    I suppose we are concerned to explain economic growth — that is the explanadum. Part of the explanation is creativity, which then calls for some explaining. So perhaps we have a structure of two-step explanation as follows:

    econ-growth — creativity — institutional conditions (liberty, etc.)

    So creativity is explanation for econ-growth, and explanadum of institutional conditions. I guess that makes sense.

    Now, is talking about these matters, as Smith, Rae, Lauderdale, Say, Schumpeter, the Swede Dahmen, Hayek, Kirzner, Baumol, Gilad, Harper, Koppl do, the same as modeling creativity?

    Well, I guess that depends on the genre meant by “modeling.” So far as Max U theorizing goes, I am inclined to say that it is not really apt to the task (as Humberto Barreto says). If you specify the the boxes searched over (for possible recombinations or whatever you want to call the new interpretations of things), you lose the feeling and sense of creativity.

    Alternatively, you might have epiphany-like discoveries coming down from a deeper, cloaked level of mind, but does that really count as “modeling creativity”? Alternatively, you can uncloak, fully articulate that second level, and even specify asymmetric knowledge within the mind. You can then say that the lower level has the experience of creativity, but it has been demystified at the higher level, which is just maximizing over a bunch of givens. (A Sept 08 AER paper models the brain as a two-level with asymmetric knowledge, and has the higher level simply maximizing.)

    Moreover, any such construct would be foolish if it failed to acknowledge that the highest articulated level has tacit contacts to the lowest still unarticulated levels (as the AER paper fails to do). The open-ended “model” — n levels articulated and tacit contact to n+1 — is the “deepself” of my 1999 Rev Aust Econ paper.

    So I’m inclined to say that Max U theorizing really doesn’t allow for the “modeling of creativity”.

  8. Bob Says:

    Thanks for the heads up on the Velupillai paper. Very interesting.

  9. koppl Says:

    Dan,

    That all seems about right to me. You can have a Max U model of creativity, but it is hard to imagine how such a model could do any real work for you.

    Bob,

    Glad you liked it! I’m quite keen on Velupillai’s “computable economics.” I think Velupillai’s 2007 result is important. I use it my paper on “computable entrepreneurship.” (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1248409)

    I am also very keen on the new results in computability that begin with

    Lewis, A.A. (1992). Some aspects of constructive mathematics that are relevant to the foundations of neoclassical mathematical economics and the theory of games. Mathematical social sciences, 24, 209–235.

    After Lewis came

    
Tsuji, M., da Costa, N.C.A. & Doria, F.A. (1998). The incompleteness of theories of games. Journal of philosophical logic, 27, 553–564

    and other works by da Costa and Doria.

    These recent results show that computability issue crop up in contexts we had thought of as “finite,” because our vague descriptions allow an infinite variety of finite games to fit the description. Thus, you get computability problems cropping up in contexts we had thought immune to them. In a forthcoming JEBO article, Prasad says, “Even for games with computable equilibrium points, best responses of the players may not be computable.”
    Computability problems are ubiquitous. I think that matters for economics. It matters for discussions of creativity. It matters for the socialist calculation debate. It matters for policy analysis.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: