“In other words, if we think of state borders as physical barriers, do we also irrationally imagine that these borders protect us in some way? …
The idea was that the dark line would reinforce the biased notion that borders are impermeable—and that states are therefore meaningful categories to rely on for decision making. …
As reported in October in the online version of the journal Psychological Science, when the radioactive waste was being stored in neighboring Nevada, residents of Salt Lake City perceived much greater risk of contamination if the border was a light, dotted line. In their minds, that light, sketchy border minimized the distinction between Utah and Nevada—and thus increased their perception of risk. The thick, dark border offered psychological protection from radioactivity.”
The use of the term “state economy”, and even more to the point, terms like “Michigan economy” and “Nevada economy” by economists, economic policy experts, and policy makers is common and has the same psychological effect as drawing a dark border around a state. The use of such terms reinforces and perpetuates the illusions that state boundaries have economic importance and that state economic policies have the power to change economic outcomes.
A state economy is nothing more than an artifact of geopolitical decisions made long ago. Like the mix of bird species in a state, the mix of economic activities and relationships in a state is little more than an arbitrary consequence of the intersection between where a state boundary was drawn and where particular economic activities and relationships later developed. Just as habitats expand and contract and change shape and location with time, so too do the economic boundaries defined by the distributions of economic activities and economic relationships among people. But, state boundaries almost always stay put.
There is now only one economy, the world economy. State economies exist only in our minds. So too, the power of state governments to improve investment and employment outcomes exists only in our minds.
Annual Change in Distribution of Percents of State Populations That Are Employed
What is striking in the first chart above (employment change for groups of states defined by Tax Climate Ranking) is that the patterns of change are so similar across the groups. One might easily imagine the changes are choreographed.
This much unity of change is not consistent with the oft stated assertion that state policies have a large impact on state level employment change. It is much more consistent with the proposition that state level employment change is driven primarily by economic forces that transcend state boundaries.
This interpretation is supported by variations within Tax Climate Ranking groups (second and third charts). If state economic policies produced large employment effects, one would expect states with very different Tax Climate Rankings to proceed along visibly different employment change paths, while states with similar Tax Climate Rankings would proceed along similar employment change paths. This isn’t the case.
The differences among the patterns of change for the closely ranked top ten states and among the patterns of change for the closely ranked bottom ten states appear to be greater than the differences among the patterns of change for groups of states.
Other employment change charts using the same formats (not included here) display very similar patterns. Again, patterns of change across groups are very similar and differences within groups are as large as differences across groups.
For those charts the same employment change data were sorted differently. One set of charts displays patterns of change for ten groups created by sorting the employment change data on percent of total population employed in 2000 (on the assumption that different initial employment levels might correspond to different policy histories and thus produce divergent change patterns). The second set displays patterns for groups created by sorting the data on change in the percent of total population employed from 2000 to 2010 (on the assumption that differences among states in how well they performed over the decade might reflect policy differences other than Tax Climate that would produce noticeable differences in patterns of change).
The last item above, the table showing the distribution of percents of state populations employed, offers more evidence that state employment levels change in unison. Over the course of the decade the highest percent of population employed and the lowest percent change together, producing an almost constant range between highest to lowest. Again, this is much more in keeping with conclusion that economic forces that transcend states in scope, not state policies, drive employment change at the state level.
A Plausible Explanation
It would be very surprising to find policy differences across states large enough to produce large differences in patterns of employment growth.
First, the states are part of a national governmental system in which constitutional provisions limit the policy options available to states, state governments are very similarly organized, and policy makers generally agree on the beneficence of the free enterprise system. These factors inhibit the development of large policy differences among states.
Second, when a state enacts a policy or set of policies that seem to provide it with investment and employment growth advantages over other states, those other states adopt those or similar policies very quickly. Thus, large policy differences will not persist.
Third, U.S. states are exposed to economic forces that are global in scope. Global financial and production corporations move large volumes of money and commodities across state and national boundaries. So do many local businesses. Ownership often transcends state and national boundaries. States are subject to the policy rules of numerous bilateral and multilateral trade and investment agreements entered into by the U.S.
These global economic forces are certainly powerful enough to overwhelm the effects of policies limited in scope to state boundaries and to wash away the employment change differences among states that policy differences might otherwise produce.