In a new paper with the National Taxpayers Union, I explore the tangible personal property (TPP) taxation of data centers and other cutting-edge, capital-intensive industries. I shared a short primer on TPP taxes a few months ago and won’t repeat that here, but I did want to draw out a few points from this new paper.
1. Personal property values are not localized.
The value of real property (land and structures) is highly location-dependent. Local services like schools, roads, public safety, and local amenities all make land more valuable. A tax on real property not only roughly approximates the value of the benefit received, but the services those taxes pay for also increase the property’s value.
None of this is true for machinery and equipment. A home in a good school district is worth more. A server rack is worth the same everywhere. Whereas additional homes and other structures can increase local government costs, the value of the machinery and equipment within a facility has little impact on local costs. High tax costs on mobile property can backfire.
2. The United States is a global outlier.
Of the 38 countries in the OECD, only seven tax business personal property. Among major economies, only the U.S. and Japan have substantial TPP taxes. As far back as 1966, one scholar called TPP taxes “the most American of property taxes,” and not in a good way.
3. TPP taxes overvalue the property of cutting-edge industries.
Valuation methods, particularly the reproduction-cost method (described in the paper), combined with long depreciation schedules, overstate the value of rapidly obsolescing equipment. Traditional manufacturing equipment can last for decades, and while it must eventually be replaced, its replacement may only modestly improve the prior equipment’s productivity. That is not true in areas where technology is advancing rapidly.
The newest AI-oriented chips represent a massive advance over those in four-year-old servers. While the new servers are extremely expensive, their gains in computing power mean the “cost of compute” is declining (Moore’s Law in action). If a TPP tax values a four-year-old server based on some trended share of its original or identical-replacement cost, rather than on its actual current value, it can dramatically overstate—and overtax—the equipment.
4. Taxes on capital are stacking up.
TPP taxes apply to data center capital every year. Meanwhile, states are increasingly reconsidering sales tax exemptions on the equipment (which I’ve written about previously), and federal proposals would delay cost recovery on those equipment purchases. That’s potentially three different layers of tax burden on capital equipment.
Lawmakers have traditionally understood that capital investment is desirable. If they abandon this idea for data centers and AI, there’s a real risk of spillovers to other cutting-edge industries as well. TPP taxes already overtax innovative capital-intensive businesses. If state and federal governments pile on further burdens, they risk seriously disincentivizing capital investment and hobbling economic growth.
Obligatory Marketing Note
My consultancy provides tax policy research, writing, and other services, both project-specific and on retainer (or in visiting fellow-style roles). If you are in the market for tax policy research or know someone who is, please let me know.
Please Share this Substack
If you find this Substack valuable, please do me a favor and share it with colleagues and others who may be interested. And if you haven’t yet subscribed (it’s free), please consider doing so!