In setting the 2015 calendar parameters for health plans and employers, Kathleen Sebelius, the secretary of health and human services, quietly did some creative but questionable arithmetic that forced taxpayers to give still more help to businesses and people who buy health insurance.

The Affordable Care Act is designed to encourage people to enroll in a health insurance plan and to shop around for value for their medical dollars.

On the first point, the law says large employers will be charged a penalty for not providing coverage to their full-time employees (and thereby helping their employees get taxpayer-subsidized coverage). The amount of the penalty was set to be economically significant relative to the costs of coverage.

On the second point, people enrolled in health plans are asked to pay part of their medical expenses – “cost-sharing,” as the law calls it. This is why the plans sold on healthcare.gov have high deductibles in comparison with traditional employer plans. The amount that plan participants are supposed to pay is supposed to be commensurate with the costs of medical care.

To achieve both of these goals, the law specifies that the cost-sharing rules and the employer penalty be indexed to health cost inflation. Specifically, the Affordable Care Act says that in each year after 2014, the employer penalty and cost-sharing parameters will exceed the value they have in 2014 by a percentage equal to the “premium adjustment percentage,” which is “the percentage (if any) by which the average per capita premium for health insurance coverage in the United States for the preceding calendar year (as estimated by the secretary no later than Oct. 1 of such preceding calendar year) exceeds such average per capita premium for 2013 (as determined by the secretary).” The “secretary” refers to the secretary of the Department of Health and Human Services, currently Ms. Sebelius.

The average per capita premium for health insurance coverage increased in 2013, especially in the individual market, because the Affordable Care Act required plans to provide more benefits. For example, the eHealth price index was about 40 percent greater during the first quarter of 2014 than it was for calendar year 2013 (see this chart, in which the dotted red line is the 2013 average). This is no surprise – more benefits mean higher premiums – and I presume that Congress understood this.

This is not to say that high-premium high-benefit plans are undesirable, just that, without subsidies, you get what you pay for, and pay for what you get.

But a political problem arises in that a premium increase that averages, say, 40 percent would require a 40 percent increase in the caps on what individuals with coverage can be asked to pay for their own medical expenses and increase the employer penalty by 40 percent (above what it would have been had it been enforced in 2014).

Among other things, the salary equivalent of the employer penalty next year, with a 40 percent premium adjustment percentage, would be almost $4,300 per employee per year.

To make matters (politically) worse, the increase in premiums from 2013 to 2014 is likely to be permanent, because the new rules on minimum benefits are permanent (as the law now stands). In other words, an increase in caps and penalties next year would be likely to last long into the future. The Department of Health and Human Services needed a way to measure the average premium for health insurance without acknowledging what is actually happening to health insurance premiums.

The department explains the two principles behind its solution. The first principle is to estimate premiums with its own projections, rather than averaging actual premiums observed in the marketplace. The second principle is to limit its use of data to market segments where “the premium trend is more stable.”

Since only one year has passed since 2013, for now that means limiting the data used to market segments where the premium adjustments are sufficiently close to zero.

In particular, for now the premiums in the individual market will be ignored for the purposes of estimating changes in premiums. As a result, the secretary has declared that the premium adjustment percentage is but a fraction of 40 percent: 4.2 percent, when rounded off.

I am not saying that the premium adjustment percentage should have been 40, just that it should have been based on actual transactions in all of the market segments, and as a result significantly greater than 4.2.

Perhaps taxpayers of the future will remember March 11, 2014, as the day when one cabinet secretary added billions of dollars to the deficit.

Casey B. Mulligan is an economics professor at the University of Chicago. He is the author of “The Redistribution Recession: How Labor Market Distortions Contracted the Economy.”