No need to insure people you don’t employ.

The recent accusations of “exaggerating costs and underestimating savings” levied on the Congressional Budget Office by White House Budget Director Peter Orszag are only the most recent salvo of an ongoing attack on the credibility of the the agency. A chorus of voices, including President Obama and many ranking Democrats in the legislature have consistently laid the charge in recent months that the CBO “always give you the worst case scenario on one initiative and never … any credit for anything that happens if you have early intervention, health care” in the words of House Speaker Nancy Pelosi.

Any projection will have flaws and oversights, and it is entirely possible that the preliminary analysis underestimates some savings, though it clearly does include them. The other possibility is that assumptions about revenue and participation are woefully optimistic and the deficit impact will be even greater than indicated.

In a letter ironically hailed by some late critics of the agency, the CBO laid out their expectations of the effects of the proposed legislation on private insurance enrolment, private sector premiums and the labor market, among other topics. Their projections are relatively supportive of the narrative advanced by the president and other Democratic leaders; certainly more so than those by private firms, such as the Lewin Group. However the letter, like previous communications and testimony, points out the difficulties and uncertainties inherent in the process.

One key factor are that may be underestimated is the impact on the labor market. Though the analysis does acknowledge that fees on employers not offering insurance “would tend to reduce the hiring of workers at or near the minimum wage, because their wages might not be able to decline by the full amount of the fee” they are dismissive of the potential impact, citing studies suggesting that moderate increases in minimum wage have little impact on employment. The relationship between minimum wage and employment is not wholly uncontroversial, though, and the compound effect of wage increases and health care tax stretch any credible definition of moderate. The nominal cost of employment will have increased by 51% over a 2007 baseline.

It doesn’t take much imagination to invent mechanisms by which small businesses might reduce their payroll in order to reduce or eliminate their obligation under the proposed legislation. Direct employees could be displaced by independent contractors or outsourcing contracts. Chains could be reorganized as independent franchise operations. Increased automation, such as self checkout machines, could eliminate positions entirely.

The extent may be arguable, but losses are certain, and the casualties will entry level.