I agree with almost everything that Steven Weinberg says about manned (bad) and unmanned (good) space exploration in http://www.nybooks.com/articles/17011, but there's a bit of economic illiteracy towards the end that made me wince:

"My training is in physics, so I hesitate to make pronouncements about economics; but it seems obvious to me that for the government to spend a dollar on public goods affects total economic activity and employment in just about the same way as for government to cut taxes by a dollar that will then be spent on private goods. The chief difference is in the kind of goods produced by the economy—public or private."

Among the things which these statements neglect are:

  1. The resources wasted in collecting taxes. This includes the budget of the IRS ($10.4 billion in FY 2003) and all the time and effort spent by people in preparing their taxes (4.6 billion hours in 2001 according to http://www.cato.org/dailys/04-15-02.html).

  2. The resources wasted by people changing their behaviour because of taxes. These are hard to measure directly but are a nontrivial percentage of the total revenue collected ($1.9 trillion collected by the IRS for 2002).

  3. The resources wasted because of the relative lack of incentive to spend its money efficiently. Again, almost impossible to calculate exactly, but certainly a nontrivial percentage of the total spent (federal spending was over $2 trillion in 2002).

  4. Perhaps most importantly, "total economic activity and employment" is (despite what you might hear politicians say during an election year) a horrible measure of the goodness of any economic policy. Employment could be maximized by forcing everyone to dig ditches and economic activity could be maximized by designing machines that allow people to hand dollar bills back and forth extremely quickly, but doing either of these things would be uncontroversally stupid. There's no one best measure of economic goodness (except goodness itself), but Econ 101 teaches any number of measures (real GDP growth, real median income growth, Pareto optimality) that are better than this.

Well, at least he hesitated before pronouncing. I guess that's something.