Which was that he would not cut any of its benefits, a promise made during the 2008 campaign and now broken with his proposed budget that will replace using CPI-W for the COLA with the chained CPI-W, the former estimated by many to rise about 0.3% more per year than the latter. However, nobody should be surprised that he has done so as he has been signaling a willingness to do so for some time now at least since the 2011 budget negotiations with Congressional Reptards over the debt ceiling increase (and, yes, he should declare the debt ceiling unconstitutional and dispense with that bloody thing). I am not going to comment on the politics of this move, although plenty of people argue that it is not a winner in any way shape or form.
As it is, I weirdly take this sort of personally. Back during the 2008 campaign, Bruce Webb and I wrote a memo to the Obama campaign at a time when he was supporting some sort of "Social Security Reform," arguing that such was not needed, and he made his promise not to do anything at all to or about SS shortly after our memo was sent, although I suspect that it was broader political forces that were responsible for his change of mind then rather than our memo. However, I took the promise to heart and am now sad to see it definitely broken (although see more below). There are many economic arguments for why this new proposal is a bad idea.
The obvious one that has been discussed by many is that it is highly likely to reduce the growth of SS benefits for seniors in an era when arguably they should be increased more due to the collapse of defined benefit private pension plans. See http://krugman.blogs.nytimes.com/2013/05/desperately-seeking-serious-approval and http://www.theatlantic.com/politics/archive/2013/04/memo-to-president-obama-expand-social-security-dont-cut-it/274728 . This latter by Steven Hill goes on to propose a major increase in benefits in a two-tier system, funded by raising the income cap and closing certain high income tax loopholes, not a bad proposal.
The other is the point that over a long period of time it appears that the cost of living for seniors has been rising more rapidly than the overall cost of living, so that the experimental CPI-E for elderly should be used after getting it into proper shape. This higher rate of elderly cost of living increase has been largely due to the greater use of medical care by the elderly, which has risen more rapidly than the general rate of inflation for a long time, and has been 0.2% greater per year between 1982 and 2011 than the CPI-W. Among those making this point previously have been Dean Baker http://www.cepr.net/index.php/blogs/cepr-blog/thoughts-on-the-chained-cpi-social-security-and-the-budget , Bruce Webb at his own blog http://socialsecuritydefender.blogspot.com , and me http://econospeak.globspot.com/2012/12/retiring-on-price-index-chain-gang , a matter that is getting more personal as I officially become a senior citizen this coming Friday.
I note that Will Wilkinson argues that between 2006 and 2012, the CPI-E actually rose by 0.1% less than the COI-W or CPI-U, although I think his argument that this reflects a deceleration of medical care cost increases does not explain it. He supports going to a chained CPI-E index, http://www.economist.com/blogs/democracyinamerica/2013/04/barack-obama-budget . In any case, if the main source of difference between the measures is medical care costs, this is all the more reason to get the rate of medical care cost increases under control and in line with the other sectors of the economy.
Having mentioned my old friend Bruce Webb's blog, I cannot avoid noting his discussion of "Rosser's Equation." This is simply the empirical observation that after a supposed "bankruptcy" of the system, as much of the media likes to call it, that is projected to happen in the 2030s if nothing is done, recipients would actually be better off in real terms than are current ones although they would experience a cut from what they had been receiving at that point, with me estimating this in 2005 at 120% better, although more recently Dean Baker has it at 125%. In any case, I must credit Dean with noting this before I did, although it has always been in the Social Security Trust Fund reports for anybody to see who looked closely enough, Anyway, I think I made more noise about it than did Dean then, although I would be fine with it being called the "Baker-Rosser Equation," :-).
All of this reminds me that in 2005 the three of us through the old MaxSpeak that preceded this blog fought long and loud and hard against Bush's campaign to partially privatize Social Security. Although I cannot say for sure, I think that we were partly responsible for bringing the lack of a need to do anything to or about Social Security to the attention of more widely read bloggers such as Paul Krugman and Mark Thoma, who brought this argument to a wider public, with Bush's campaign failing pathetically. However, this time the threat looks more serious, coming as it does from a Dem president. The public does not want this, and it is not needed for budgetary reasons in a world of rapidly falling deficits and still stagnating employment growth, but we need to move again to make our voices heard on what a bad idea this is all around. Obama should go back to keeping the promise he made in his 2008 campaign.