n.V1-7.18 | ENTERPRISE ZONES AND FIRM SUBSIDIES
01| Draw a boundary around a poor area and cut taxes for firms that locate inside it. Jobs follow capital, capital follows incentives, and the residents get work without moving. It is the most politically durable place-based instrument in existence — visible, announceable, cheap to legislate — and it has the weakest record in this book part.
02| Neumark and Simpson's survey of place-based policies covers enterprise zones, EU structural funds and industrial cluster policies. On enterprise zones specifically their summary is that the evidence is decidedly mixed, and their honest conclusion is broader and worse: the field does not know which features make such policies work, who gains and who loses from them, or how to reconcile findings that point in different directions.
03| That is not the same as saying they fail. It is saying that after decades of use and a substantial literature, the basic questions are open. A policy instrument in continuous political favour, of which nobody can specify the conditions for success, is being chosen for reasons other than evidence.
04| The failure modes that the literature does identify are worth naming, because they recur wherever subsidies are attached to geography. Firms relocate across the boundary rather than expand, so the zone's gain is a neighbouring area's loss and the national effect is zero. Firms that would have located there anyway collect the subsidy. Jobs created go to commuters rather than residents, because the boundary constrains where the firm sits and not who it hires. And the benefit is capitalised into land values, so property owners collect it — who in a poor area are frequently not the poor.
05| That last mechanism deserves more weight than it usually gets. Place-based benefit accrues to whoever holds the fixed asset. Improve an area and rents rise; the tenant pays the improvement to the landlord and the resident who owns nothing captures none of it. Any intervention that raises the desirability of a poor place without addressing who owns it has a distributional problem built into its physics, not into its implementation.
06| Neumark and Simpson close on the question that any place-based approach has to answer, and it is the right one: what creates self-sustaining longer-run gains, as opposed to activity that stops when the money stops. A tax incentive is by construction the second thing. The subsidy is the reason the firm is there; withdraw it and the reason is withdrawn. Nothing has been built that outlives the incentive, because the incentive was the whole intervention.
07| The comparison the next chapter makes is the useful one. Within the same literature, and by the same authors, a different class of place-based spending shows a better record — and the difference between them is not about geography at all.
08| Open: essentially everything. Which features matter, who gains, why studies disagree. Neumark and Simpson say so directly, which is more than the policy's advocates do.
