Friday, September 14, 2012

Arggh!!!! How Hard is it to Look Up Facts People?!?!

I try to read news from both sides. But sometimes I get a little explosion of frustration over the unexamined assumptions that pass for conventional wisdom. Today's discontent comes from reading the National Review, particularly this piece of drivel from Thomas Sowell.

I've covered at some length the fact that there is no evidence that government assistance creates dependency. If anything, the evidence is that state supports, by increasing the rewards from work, increase employment and activity in the market sector (not surprising, since people respond more strongly to losing what they have than loss of potential gains, providing a floor, and thus something to lose, then increasing upside gains is more likely to be successful given people's psychological responses than trying to beat them until morale improves since people at rock bottom tend to just grow resigned and stay there, despite popular movie stories to the contrary).

But, there is a linked idea that receiving government handouts increase voting for the Democrats. However, a quick search for ungated papers on the subject reveals that receipt of means tested benefits (if an explanation is necessary, aid targeted at the poor rather than middle class) decreases civic participation and voting. Since assistance targeted at the poor has a stigmatizing effect, it would be a rather boneheaded move to target programs towards the poor if the goal was electoral success. While one is free to take this as another problem with the welfare state, it does contradict the idea that hand outs to the poor are meant to secure electoral success.

Sorry, but there must be another explanation.


  1. Something that you might be inclined to agree with:

  2. Good article, but I disagree with the bits about sexual ethic, traditional family structure, and the emphasis on religious institutions to replace the state. My reasons for this are that plenty of countries have adapted to changing family structure by adopting institutions intended to stabilize cohabitation, whereas we have adapted policies that are a mess of pushing/pulling towards/away from marriage and stable cohabitation, with incentives that shift throughout the lifecycle; all with the intent of reinforcing marriage (unlike some philosophical stances, I don't give points for intention).

    Regarding religious institutions as alternatives to state spending, I'd not that this preference rests on slim data. It is true that civil society often delivers better outcomes than government programs in the current environment, however, this does not imply this approach can be scaled up. Studies of the functioning of civil societies in weaker states rather suggest the opposite, civil society supports fail to provide individuals with needed security. The reason for this is simple, social problems need a baseline to address, once this is provided it becomes easy for motivated individuals or groups to create better solutions on a local level. However, given heterogeneity of both preferences and abilities, the availability of the dedicated, skilled personnel needed for civic groups to deliver services efficiently is insufficient to the need. The state can partially make up for this by higher spending, this can't be done at the civil society level because signs of failure are self-reinforcing. In general, civil society is superior because the most motivated individuals choose to participate, there is no reason to believe that there are more of these individuals that would move into the field if the state ceded space to civic groups (I can't help mentioning this whenever I see compassionate conservativism mentioned, it looks like a magic asterisk to me, ignoring data we have on the functioning of social support systems in states with weak welfare states, even though these states are third world there is no reason to think results here would differ).

    I'd also add that I disagree that giving more cash wouldn't help. Given international comparisons, the key thing is to give assistance in forms that are not stigmatizing, like universal healthcare or good public transportation. Non-means tested programs tend to have an outsize impact on the lives of the poor and their future chances while avoiding the disincentive effects of means tested benefits.

    So I guess I actually disagree with a lot of it, after thinking about it. I did like the read though.

    I do agree strongly that it's primarily an institutional problem, but I see the institutional problem as one primarily resulting from stigmatizing regulations meant to weed out the undeserving from the deserving poor. I think it would go a very long way to reduce the social disruption lamented in the article by simply adapting our institutions to be non-stigmatizing, eliminating the implicit penalties that come with means testing, and adjusting our regulations to work with the way people do live and do have their families rather than trying to use our regulatory system to make them live the way we want them to and have their families the way we want them to have them.

  3. Thank you very much Tzi for your thoughtful reply. I do concur with your assessment.