Tyler Cowen seems to have gotten a lot of attention with a stimulating post. Matt Yglesias and Brad Delong have responded so I suppose I really don't have much useful to add.

I'm not going to be as polite as Yglesias and especially Brad, so I'd like to thank Cowen now for a very interesting post and for opening an interesting discussion.

Tyler Cowen ...

" To draw another contrast with (some) libertarians, I don't believe that additional immigration is necessarily a win-win game at all margins. More immigrants will bring some very real fiscal burdens, ask anyone in California, or any hospital near the Mexican border. So if we want more immigrants, at some point it will cost us something. Furthermore the bigger the welfare state, the more the costs of immigration are socialized in an unfair, unsustainable, and undesirable way. So immigration and the welfare state are substitutes at the relevant margin. I choose immigration.

...

So that is a significant reason why I am not a liberal. I prefer high growth, minimum domestic transfers, and a higher rate of immigration. Growth plus resource mobility is the best anti-poverty strategy we are likely to find. And this recipe is closer to classical liberalism than to modern liberalism. I might also add that the United States, through immigration, satisfies the Rawlsian formula better than does Western Europe."

I agree with Cowan that restrictions on immigration are arbitrary and unjust. I think a just society would let anyone who wished immigrate. The only limits I would accept are a cowardly surrender to violent natavists who would get violent if there were too much immigration. My target immigration for say the USA is at least 10 million a year (still piddling compared to the problem but getting close the practical limit). By the way, I wonder if Cowan accepts pragmatic limits or wants totally free immigration ?

My problem is that Cowan's conclusion that he is not a liberal rests entirely on his unsupported assertion that "So immigration and the welfare state are substitutes at the relevant margin." I am sure Cowan does not make the case, but I am not sure what his assertion means. I will try

1: the trade off is fiscal I. Given current laws and programs, immigration would make the welfare system fiscally unsustainable.

As noted by Yglesias, this claim is based on a innacurate implicit description of existing programs. The main social welfare programs in dollars are old age pensions and medicare and immigrants are generally young. Immigration would help make current programs more nearly sustainable. The costs of increased welfare (tanf) food stamps and schools would be more than balanced. In (literally) old Europe, it is widely argued that immigration is needed or else the social welfare system will be unsustainable (of course it will be in any case). This is argued by everyone who has looked at the numbers. Europeans restrict immigration in spite of the problems with their welfare system not because of them.

2: The trade off is fiscal II. It would be impossible to provide the sort of generous social welfare that liberals want and have free (or massively increased) immigration.

First note above, that would have to be well beyond European levels of generosity.

Second it would be possible to have neither. Egalitarians do not have to choose between advocating more generous social welfare programs and advocating free(er) immigration. In no case will we win the political battles so thoroughly that we will hit the social budget constraint. The fiscal implications of athe combination politically impossible level of immigration and a politically impossible generosity of social welfare programs is not "relevant" to the real world.

Notice I am arguing about the sign of the interaction effect of immigration on the affordability of social welfare. At current and plausible future levels of welfare generosity, immigration and social welfare are fiscal complements not substitutes.

3. The trade off is political. We have to decide whether to fight for more immigration or to fight for more social welfare.

This sure does not support a libertarian position, which would imply fighting for less social welfare. More to the point, as noted by Brad, this is just not true for voters or campaign donors. I think that generally pro immigration politicians are also pro social welfare. I am sure that there is not a strong pattern which generally forces voters to choose one or the other.

(Humerous ?) Case in point G.W. Bush proposed a relative loosening of restrictions on immigration and forced through the biggest new social welfare program in generations.

4: The trade off is a rhetorical trick. Works for me.

Even if Yglesias and I are wrong, notice an anti quantitative trick. The argument is
I. there is a (one to one) trade off between immigration and social welfare programs
II. (One unit of ) immigration does more for the poor than (one unit of) social welfare programs.
III. Egalitarians should be against social welfare programs.

The conclusion requires the statements including the words in parentheses. However, the statements (with the words in parentheses) are not demonstrated. In fact, they are not even stated.

How about this version

A. For each addition twenty units of social welfare programs immigration is reduced one unit
B. One unit of immigration does ten times as much for the poor as one unit of social welfare
C Egalistarians should be for social welfare programs

ABC is as consistent with Cowen's argument as I.II.III except, of course, for the conclusion.