Costs of a Basic Income Guarantee (Bleg)

I have supported the basic income guarantee (BIG) before. Recently, when researching for an article I wanted to write, I came to doubt whether BIG is really the “second best” welfare program many libertarians, and others, make it out to be. Now, I prefer to label myself as an agnostic, because I haven’t done the math.

In any case, Steve Horwitz, in his Reddit Ask Me Anything, writes,

If nothing else, [BIG] has far less overhead cost to taxpayers and gets rid of many of the perverse incentives of the current welfare system, especially with respect to marriage and families.

He’s talking about two sources of benefits,

  1. Less direct cost to the taxpayer;
  2. Efficiency gains from removing perverse incentives created by other welfare programs (which would, ideally, be removed).

Is (1) right? Here’s Danny Vinik, at Business Insider,

In 2012, there were 179 million Americans between the ages of 21 and 65 (when Social Security would kick in). The poverty line was $11,945. Thus, giving each working-age American a basic income equal to the poverty line would cost $2.14 trillion. For some comparison, U.S. GDP was almost $16 trillion in 2012 and the defense budget was $700 billion.

According to the same article, the Federal government currently spends about $750 billion/yr. on welfare payments to low-income Americans. Even if we offered them only half of a poverty-line income, we’re still talking about over one trillion dollars per year. That is still a net increase in the direct costs of BIG to the taxpayer.

Now, we can make the case that part, or maybe all, of the difference (between the cost of BIG and current welfare costs) would be made up by an efficiency gain. But, that’s an empirical claim. Has anyone tried to look into it?

5 thoughts on “Costs of a Basic Income Guarantee (Bleg)

  1. Becky Hargrove

    Unfortunately, welfare systems support the public/private intermediary forms of services which have grown up around them. To be sure, welfare systems need to be discarded. Except doing so with only a backup plan of basic income, just exposes services systems for the extreme inefficiencies they create through knowledge use limitations. Worse, many services in their present formation would contract to be applicable for only a small percentage of the population, even though everyone needs services product. Better to get rid of welfare when populations can create their services without knowledge use limitations.

    Reply
      1. Becky Hargrove

        Healthcare in particular, because it can be difficult for anyone with chronic illness to do an “end run” around healthcare costs. For instance, even with present day Medicare plus additional insurance options, the expenses that some patients incur in the last year of life (i.e. a couple of months in the hospital) will still mean at least $20,000 out of pocket. If all healthcare was made more local through information transfer (fast broadband in rural areas), that could cut costs.

        Reply
  2. John Snow

    I think you’re forgetting that many proponents of BIG don’t want ALL Americans to receive $12,000. Only Americans who earn anything under the poverty line. So multiply the number of Americans making under $12,000 times $12,000 and you get a very different number. Plus you’d have to calculate things like families of 4 would only get about $24,000 (not $48,000) etc. So really, the average person under the poverty line would receive less than $12,000 anyway.

    I haven’t seen any studies that show what BIG proponents (or opponents) would estimate the cost to be, but it would certainly be much much less than your figure.

    Reply
    1. JCatalan

      I think most people support a universal BIG, because giving the BIG to certain people only creates a big distortion (think 100% marginal tax rate).

      Reply

Leave a Reply