4 comments on “Timmy elsewhere

  1. Tim – presumably in effect a restatement of the same argument you have made a million times about tax incidence?

  2. “…dynamic stochastic occupational choice model with heterogeneous agents and evaluate the impact of a potential reduction in the corporate income tax on employment…”

    This looks like a variant of BiG’s second law of statistical bullshit (“We used the Vladiskov-Femangov least-squares binomial regression rank correlation exact test with Watson’s log-transformation correction” Translation: “We finally found an obscure test that gave us P<0.05")

    What happens if you evaluate the impact of a potential increase in the corporate income tax on employment using a static predeterminate fixed occupation model with homogeneous agents?

  3. Tim’s wrong: the research he cites tells us nothing about the effects of corporation tax in the UK.

    According to the paper, most of the benefits of reducing or eliminating corporation tax would come from firms changing from pass-through businesses to “C corporations”, because that would free them from restrictions on capital raising: “a pass-through business cannot have more than 100 shareholders, nor can it have any foreign, institutional or corporate shareholders”.

    Whatever the merits of that argument in the USA, it doesn’t apply in the UK where this sort of pass-through business doesn’t exist, and the tax structure already makes incorporation attractive.

    Even in the USA, the proposition looks doubtful. The authors seem not to be aware that the restrictions they list apply to “S corporations”, but not to Limited Liability Companies, and it’s LLCs which have become increasingly prevalent in the last few years – Chrysler for example is now an LLC.

  4. @BIG

    When economists build models the default starting position might be to build the simplest, but it’s not the most realistic. A result is actually more interesting if it applies in a model with heterogeneous rather than homogeneous agents, because the heterogeneity is more lifelike (so seems to make it more plausible that the model is applicable in reality).

    Of course modelling is not a perfect way of getting round the problem of the impossibility of economic experiments (ceteris paribus is a very powerful restriction on experimentation) but what you cited is not an example of an academic pulling the BS lever.

Leave a Reply

Name and email are required. Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.