My interpretation of their data: inconclusive. It is
very important to look at the left end of the graph when comparing percentages of growth. A quantity's growth of 5% each year for five years is greater than an equal quantity's growth of 2% each year for 10 years. This is what we have in this document, although the authors do not address it in any detail. The graph on page 9* shows that the non-minimum wage states had higher growth rates in the first five years of the graph, which is much more important than the non-minimum wage states' superior growth in the last five years of the graph.
Indeed, the non-minimum wage states posted a cumulative nonfarm job growth of 10.2% from 1991 to 1996, versus 2.8% in the minimum-wage states. This shows us at least two things: it now became more difficult for the non-minimum wage states to maintain these rates of growth (it is harder to grow 1,000 by 10% than to do the same with 100). This early growth also provided a large base which would allow the non-minimum wage states to outperform the minimum wage states throughout the remainder of the graph, even when they were periodically surpassed by the growth among minimum wage states.
As a result, nonfarm employment in the non-minimum wage states had grown by 23% from Jan 91 to Jan 05*, versus 17.1% in the minimum wage states.
Though it appears that the non-minimum wage states outperformed the minimum wage states, I can not draw a conclusion about the effect of minimum wage. There might have been an economic crash prior to 1991, which may have disproportionately affected the non-minimum wage states, thus allowing for the comparatively high percentages of growth from 1991 to 1996. There may also be a myriad of other factors at work here.
I do, however, take issue with the premise of the document. Upon close examination, one will find that all of the statistics are derived from 1998 onwards, which FPI says reflects the introduction of state minimum wages. Although that analysis is valid within that context, there remains the glaring omission of the comparatively low performance by the minimum wage states-to-be in the early 90s. At the very least, this demonstrates that there is a difference in performance that is unrelated to state minimum wage. Yet when the strong growth of the non-minimum wage states
is discussed, it is subtly connected with the presence of the federal minimum wage.
*Note that the graph in question appears on page 9 of the document itself (not the pdf page number). I used Jan 05 as a stopping point because the graph does not have a complete set of data spanning to Jan 06. The impact of the omission of the data beyond Jan 05 is probably negligible, as there is a difference of 5% to be made up by the minimum wage states.