Friday, April 1, 2011

More on House GOP Cutting Food Stamps Instead of Farm Subsidies

12 comments:

TJE said...

Some think food stamps are paternalistic and that we should just give eligible people the cash. Others think that food stamps should not be used for junk food:

http://www.allvoices.com/contributed-news/8507058-should-the-government-legislate-what-lowincome-people-on-food-stamps-should-eat

What do PBM, Mr. L, Ms C and others think?

PL said...

On the paternalistic food assistance vs. cash payments, the literature finds that there is essentially no difference in reality. The value of food assistance does not exceed the food needs of the vast majority of families, so there are very few families and individuals who receive food assistance now that would react differently if you gave them cash (and many of those families would only have that option because they are able to also use charitable food sources and other non-government, non-labor market sources of income and assistance to acquire enough food, not because they are somehow stingier or more efficient with their food consumption.) Basically, you could give people cash, but nearly all of it would get spent the same way. So there’s no significant real-world benefit to changing it from SNAP to cash-payments.

But I guess people concerned about paternalism might take that fact and argue that we should just give them cash (even though it makes little difference in outcomes).I find this line of argument against paternalism very troubling, because it intentionally or unintentionally plays into most Americans’ irrationalities, biases, and lack of knowledge about government programs. Moving SNAP assistance to a cash-support program would almost certainly lead to a major decline in assistance provided low-income households (almost 90% of all households receiving SNAP are below the poverty-line). In my opinion, this is the true goal of many people making noise about paternalism- especially when you consider that oftentimes these are the same people arguing that poor people just need to get married more often and have less children out-of-wedlock. Talk about a paternalistic approach!

Why would moving SNAP to a cash-support alternative result in reduced assistance? Unfortunately, Americans are far less supportive of cash support programs, and politicians and those opposed to government assistance of any kind can take advantage of this to further erode America’s safety net. A great example of this phenomenon is Reagan’s outrageous stories about “welfare queens” that he spread for years, even as every media source investigated his stories and found they had no relation to reality. But it didn’t matter- people still believed those claims about cash-based assistance programs which fed into their racial and moral-biases. It’s much harder to make those attacks, though Republicans still try to with their Welfare Reform Act of 2011 rhetoric, against a program like SNAP where the government support clearly and directly goes to providing a basic need, food, to low-income Americans (the vast number of who are children, elderly or disabled). With cash assistance the money might have the same indirect effects, but it’s much harder to educate the American people on that. Instead politicians would just use the very rare and worst incidents to play off Americans’ well-established racial stereotypes associated with “welfare” programs. Where I come from, the word welfare is still used as an epithet, and I know that phenomenon is unfortunately true for most of the country.

PL said...

Most Americans strongly support various forms of assistance like SNAP to low-income Americans, but they have very mixed feelings about cash/welfare programs- even though they have the same impact. If you don’t support robust government assistance programs or you don’t think the government should try to reduce poverty by promoting better life choices (i.e. getting a HS degree, married, and not having out-of-wedlock children), then it’s probably reasonable for you to believe that excessive paternalism is a good argument against food assistance. I just don’t buy into either of those positions.

The second argument “Others think that food stamps should not be used for junk food” is an interesting follow-up because it’s an example of paternalist argument- so this one contradicts the first anti-paternalism argument. Quick note: The link cited makes the mistake of thinking food assistance can be used at restaurants for prepared food- it can’t. The only exception to this is very limited/monitored programs provided in a few regions for homeless and disabled people- who shockingly have trouble preparing food!

So requiring participants to prepare their food provides a limitation on unhealthy eating habits. Also, it’s very hard to structure a program to eliminate unhealthy foods- imagine the bureaucratic nightmare of determining which foods were healthy and which weren’t- and then enforcing those rules. Also, nearly every food is acceptable or even healthy in moderation- would you try to take that into account when structuring such a program. “You can by 10 lbs. of beef from Brand X in a month, or 8 lbs of beef from Company Y because it’s less healthy, but you can only have six ounces of coffee from Company Z.” I don’t see this as being a very practical policy option.

Also, SNAP already takes a less paternalist approach to achieving this goal (have SNAP participants use the support to buy healthy food) by requiring all states to operate nutrition education programs that help participants make healthier food choices. I’d also be concerned about this change to SNAP because many program participants also live in food deserts where it can be hard to find healthy food options, and what little there is far more expensive than what most middle-class families face when buying healthy food options like produce.

TJE said...

http://yy9g.info/AmberWaves/feb03/Findings/cashnotfoodstamps.htm

PL said...

The problem with that figure is that it's wrongly using averages and aggregate results to lead to bad impressions that could lead to bad policy prescriptions. I made this point in my first comment. It's not that all, or even most or many, households would make different decisions if given cash assistance over food assistance because they spend more on food than they would otherwise. It's that some, a tiny fraction of all households, would make different decisions if given cash assistance (partly because they are able to take advantage of non-market food sources like food bank assistance.) So very few households would make different decisions if given cash, which doesn’t even speak to whether they’d be better off.

Households benefit from being given a strong subsidy for purchasing an adequately nutritious diet (after all, nutritious foods cost a lot more than unhealthy foods, so with more money they are more able to purchase healthy foods like produce). Just because some households would spend less on food if given cash assistance doesn't mean that's a better outcome for these families, or for the government. After all, SNAP is based on the USDA's Thrifty Food Plan, which seeks to provide a nutritious diet at a minimal cost. This compares to the other three more generous USDA food plans it develops based on food prices and nutrition information- the low-cost, moderate-cost, and liberal food plans. I'd be concerned that families spending less on food than what SNAP benefits provide for would be shooting themselves (and their children) in the foot by creating long-term health risks and hurting their ability to engage in the labor market. Low-skill jobs in the service sector, which many low-income people depend on, are very physically demanding so to maintain enough productivity and work stamina so it's important these workers and families get some basic level of nutrition- especially since work requirements force many into the labor market.

We also have to consider that many children benefit from food assistance, so it's important that government programs provide support for ensuring these powerless individuals (how often can they make their own food selection decisions?) have a decent chance of getting adequate nutrition. And SNAP has actually been shown to have these positive health impacts. For instance, the Food Stamp Program has been shown to increase birth-weights, reduce premature births, and improve infant mortality (for some of the relevant literature, see Almond, Hoynes, and Schanzenbach 2007).

TJE said...

http://www.fns.usda.gov/ora/menu/Published/snap/FILES/ProgramDesign/Cash-out_Administrative.htm

PL said...

I’ll assume with the link you weren’t trying to highlight the issuance costs, theft and fraud concerns with the OLD food assistance program- since these concerns have all been largely addressed with the numerous reforms the program has gone through to reach its current existence as SNAP, the Supplemental Nutrition Assistance Program. The program now operates using Electronic Benefit Transfer (EBT) cards and as a result has seen steadily declining administrative costs and error rates for years.

I’m assuming the point of the link was to continue our line of discussion and highlight the study’s finding that “There is reasonably clear evidence that cash-out will reduce expenditures on food: across three different sites, food spending fell roughly 5 to 20 percent when food stamp benefits were converted from coupons to check.” This by implication would support your position (evidenced by your AmberWaves link) that there are differences in providing these benefits via cash support or the in-kind benefits of food assistance.

In response, I’ll turn to the paper “Consumption Responses to In-Kind Transfers: Evidence from the Introduction of the Food Stamp Program,” by Hilary W. Hoynes and Diane Whitmore Schanzenbach. October 2009. American Economic Journal: Applied Economics, 1(4), p. 109-139. In a weird coincidence, this paper was used by both Professors Pliskin (when substituting for Professor Jensen in Microtheory) and Professor Hagstrom (in Economics of Poverty) when lecturing on this topic:

From the conclusion, “Economic theory has strong predictions about how consumers will reallocate their spending in response to in-kind transfers. Despite the well-known theoretical predictions, there has been relatively little empirical work to test those predictions. Our results, while mixed in statistical significance, are uniformly consistent with theoretical predictions. First, the poor react to in-kind transfers by reducing their out-of-pocket spending on the targeted good (although these results are not statistically significant). Second, total consumption of the targeted good from all sources (cash outlays and in-kind transfers) increases. Third, providing food stamp benefits in voucher form leads to a minimal distortion of the consumption choice relative to what it would be if the benefit were provided in cash.”

PL said...

Another way to consider the issue is to look at the benefits provided and realize how hard it would be for someone's food consumption to not exceed that amount naturally. The maximum monthly benefits a family of four could receive is $668 (an temporarily enhanced amount only accomplished through the Recovery Act in response to the recession), which divided by days (30), family members (4), and meals (3- assuming you don't eat at any other time?) works out to $1.85 per person per meal. People in SNAP clearly aren't be showered in excessive food assistance.

TJE said...

But why then not let people make their own decisions? Is it mostly a matter of political calculation?

And by your logic, why not require SNAP to be used only for healthy foods? I imagine that it would be a "snap" to program cards to reject soda, twinkies, etc.

TJE said...

Just sayin, Mr. L

http://www.cato-at-liberty.org/public-transit-a-classic-example-of-government-in-action/

PL said...

http://www.fns.usda.gov/ora/menu/Published/SNAP/FILES/ProgramOperations/FSPFoodRestrictions.pdf

http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VCB-4V2PT0T-3&_user=10&_coverDate=04%2F30%2F2009&_rdoc=1&_fmt=high&_orig=browse&_origin=browse&_sort=d&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=729a9878387240d9918e0557a68f63dd (unfortunately I haven't found a way through hamilton's research tools or what's available on the internet to access the full article, though the abstract gives you a good idea of their results).

Here's a powerpoint presentation on this paper: http://aic.ucdavis.edu/obesity/pubs/Alston,etal_July2007.pdf (hopefully it loads for you, I had some trouble with it. Pgs. 15-18 provide the study's results).

In case the pdf slides don’t load, here's what FRAC had to say about the study's results:
“Researchers at the University of California-Davis explored the potential impact of limiting SNAP/Food Stamp purchases to healthy foods such as fruits, vegetables, whole grains, lean meat and fish, low-fat milk, and water. They concluded that while such purchasing restrictions may improve diets and reduce obesity among some SNAP/Food Stamp participants, a number of potential consequences make it an "ineffective and inefficient" strategy on its own. For example, these purchasing restrictions may diminish the perceived value and attractiveness of the program, causing a decline in SNAP/Food Stamp participation. Eligible, "healthy" foods with their higher demand could increase in price, while ineligible, "unhealthy" foods with their lower demand could drop in price. As a result, healthy food consumption may increase among SNAP/Food Stamp participants, but decline among non-participants, especially low-income non-participants who cannot afford the higher-priced items. Updates to the list of eligible foods could also increase SNAP/Food Stamp administrative costs, given the continuous stream of new and reformulated products on grocery store shelves. Finally, the impact on unhealthy food purchases and consumption is unknown because ineligible foods could still be purchased using other payment methods."

PL said...

As I've said before, I think there are serious implementation challenges to this idea, and I'm also concerned about the food deserts and higher healthy food prices faced by many in low-income communities. There are better policy options for addressing this concern.

On food assistance vs. cash benefit debate, yes- it's largely about political concerns. After all, I've amply demonstrated that this change would have no material benefit for nearly all SNAP households, so it's not like SNAP participants would be better off.

It's not that anti-poverty advocates don't want a near-universal cash-based entitlement program (which is what this change would create). This of course is the last thing conservatives’ want- so I find this debate a little disingenuous because conservatives have clearly expressed their opposition to these types of programs. It seems quite clear to me that many (I’m not saying you) who advocate for this change are really hoping to weaken the program’s political defenses in order to enable them to reduce the program’s enrollment or benefits.

The structuring of a program is important to establishing its political viability- programs for the poor often turn into poor programs. History has shown us that to protect these important anti-poverty programs, it’s vital to structure them in such a way that prevents the American people from buying into the vicious slander and lies of politicians like Ronald Reagan.