Good policies will usually not “fix everything”

One surprising reaction to the Kalamazoo Promise has been to try to downplay the Promise’s success by emphasizing that many problems remain in Kalamazoo despite the Promise. While this is true, it is irrelevant to whether the Kalamazoo Promise is a good policy. Policies can have benefits far greater than costs without fixing all problems.

As an example of such a reaction, an op-ed by Richard Reeves of the Brookings Institution appeared in the Los Angeles Times on July 13, 2013.  The op-ed was entitled “Free college? It doesn’t fix everything”. The op-ed informed readers that the “The Promise has lifted college completion rates, but quite modestly, and far from equally.” To support this statement, the op-ed did not cite evidence on the actual effects of the program (e.g., how is the world WITH the Promise estimated to differ from the counterfactual world without the Promise). Rather, the op-ed cited some low college enrollment and graduation rates of black students after the Promise, without any evidence presented on how the Promise might have altered such statistics.

The op-ed concludes that “The lesson of the Kalamazoo Promise is that even dramatic reductions in the cost of college have modest results in terms of leveling the playing field.”

Here is what the evidence actually shows on the effects of the Promise, based on the recent report I wrote with my colleagues Brad Hershbein and Marta Lachowska. Our report evaluated the Promise by seeking to compare how the Promise had changed educational attainment for similar students over time.

  • The Promise increases post-secondary credential attainment rates as of 6 years after high school graduation by one-third. The baseline rate was 36%, and the Promise’s effect was to increase the post-secondary credential attainment rate by 12%, an increase of one-third. Of this increase in post-secondary credential attainment, four-fifths was due to an increase in bachelor degrees.
  • Is this effect a “modest result”? That depends what you mean by “modest”, a quite subjective term. What is true is that if one compares the present value of the expected future career earnings due to the Promise’s effects on credential attainment, with the present value of the costs of the Promise’s scholarship dollars, the ratio is over 4 to 1. I would call that a large benefit-cost ratio. Furthermore, this benefit-cost ratio ignores many other benefits of the Promise. For example, for Promise recipients who would have completed college anyway, the Promise lowers post-college debt loads, which might well have significant future benefits. In sum, effects of the Promise are “large”, not “modest”, in the operational sense that the Promise seems to easily pass a benefit-cost test.
  • Does the Promise help level the playing field? That’s somewhat hard to pin down because the sample sizes for different sub-groups of Promise recipients are small enough that most of the estimated differences across sub-groups are statistically insignificant. However, the point estimates suggest that the Promise’s benefits are quite broad. For example, the point estimates suggest that the absolute effect of the Promise on the proportion of non-white students getting a bachelor’s degree is greater than its effects on white students. The implied PERCENTAGE effects of the Promise on non-white students are almost 50%, that is for every 2 non-white students getting a bachelor’s degree in a world without the Promise, there is now a third non-white student joining them in also getting a bachelor’s degree. Overall, I suspect that at least among high school graduates, the Promise probably has a greater percentage effect on future earnings for non-white graduates than for white graduates. We are doing further work with our data to explore some of the distributional effects. However, given the lower baseline educational attainment and earnings of non-whites compared to whites, it seems likely that the PERCENTAGE effects of the Promise on educational attainment and economic status are greater for non-whites than for whites. The Promise probably helps make the playing field more level.
  • Does the Promise “fix everything”? This is a straw man that, as far as I can tell, is non-existent or at least rare. I have not seen any specific quote in which someone says that the Promise “fixes everything”. No sensible person believes that a free college program will eliminate educational attainment gaps across racial or income groups, or eliminate income inequality, or increase Americans’ skills to where they probably need to be.

Why does Richard Reeves want to emphasize that the glass is half-empty? One clue is his final paragraph, where he argues that “the weaknesses in the U.S. higher education system run much deeper than financial affordability.” He then refers to the lack of high-quality vocational learning, problems with quality control in higher education, and other issues. The underlying thinking seems to be a concern that if the Promise is seen as successful, this will lessen the interest of policymakers and the public in other needed educational reforms.

However, empirically it seems that the Promise and its successes has encouraged other policy solutions rather than discouraged them. When people see hope that at least part of a problem can be addressed, they are encouraged to try to address other parts of the problem. For example, in Kalamazoo, the Promise has led to efforts to significantly upgrade after-school programs and other community supports for skills development. The Promise has also led to efforts at the local community college to upgrade support services for students.

We said the following in the concluding paragraph in our paper:

“…The Promise effects have the potential for solving only a portion of America’s skills challenge. The Promise increases postsecondary credential attainment at six years after high school graduation from 36 percent to 48 percent. Presumably some of the remaining 52 percent might benefit from receipt of a postsecondary educational credential. As one might expect, “free” college is insufficient by itself to ensure higher skill levels through postsecondary education. Other policies prior to age 18 are likely needed to improve outcomes for more students. However, simple and generous scholarship programs have the potential of being a cost-effective component of the policy toolbox to increase the educational attainment of American students. “

In other words: The Promise has a large bang for the buck, so simple but generous college scholarship programs should be celebrated, not denigrated for not “fixing everything.” However, more needs to be done. But the need for more should not be promoted by downplaying the “good news” of the Promise’s success.

About timbartik

Tim Bartik is a senior economist at the Upjohn Institute for Employment Research, a non-profit and non-partisan research organization in Kalamazoo, Michigan. His research specializes in state and local economic development policies and local labor markets.
This entry was posted in Distribution of benefits. Bookmark the permalink.