For many years, some observers have argued that Social Security cost-of-living increases are too generous. They argue that the public fisc could save billions of dollars by using a "chained" consumer price index (CPI). The idea is that CPI increases should not be based on increases in the cost of a static market basket of goods because as the price of a product rises, consumers tend to substitute lower-priced products, use less of the product, stop using the product, etc.
In this LA Times article, Michael Hiltzik says that the chained CPI makes no sense for social security benefits for several reasons, one of which seems quite important: For the kinds of things that senior citizens need -- medical care, for instance -- it is often not reasonable to assume that there is a substitute product or that the person can do with less of it. (As Hiltzik puts it, "it's not as though you can forgo a prescribed heart bypass operation and opt for a cheaper hernia operation instead.") Moreover, the CPI underestimates inflation for seniors because medical care is a disproportionate part of their spending, and medical care "has risen in cost at nearly twice the rate of overall inflation over the last couple of decades." So, at the very least, you wouldn't want to go with a chained CPI until you fixed the CPI for seniors. Hiltzik's article talks about efforts to do that as well.


[O]n calorie disclosures, I think it's important to note that some of this retooling is hopefully going on. Although not in the context of menu labeling, the FDA's announcement in October 2009 that the agency is considering different forms of front-of-package labeling gives some hope that we may see more effective forms of calorie disclosures in grocery stores and elsewhere (http://www.fda.gov/downloads/NewsEvents/Newsroom/MediaTranscripts/UCM187809.pdf). One promising model is the "traffic light" system, using the familial colors of the a traffic light to express relative levels of fats, sugar, and sodium. Sadly, while the British Food Standards Agency advised the use of the traffic light system, the EU recently rejected making the system mandatory. A less consumer-friendly model is the Grocery Manufacturer Association's "Nutrition Keys" model, which fails to utilize any effective signals for calorie information (and would only require information on saturated fat, not overall fat), and has been criticized as an attempt to circumvent more effective government regulation.
Colin's comment reminds me of a recent post by Jeff Sovern, who noted that, although most disclosures are ineffective, disclosures may work if they are simple and easy to understand.