Almost exactly 15 years ago, in September 2008, “my” side lost an attempt to rein in state rent-seeking efforts. It was the time when the decision was handed down in the deservedly obscure case, “Munger v. Google.”
Okay, it was Munger v. State of North Carolina. I have nothing against Google, I suppose. One can’t blame dogs for eating out of the garbage. It’s what dogs do. But there is no reason we have to scatter garbage on the ground to attract packs of them.
Google’s parent company, Alphabet, has a total value of well over $1.5 trillion, and has been higher. The number of employees has more than tripled since 2015, rising to more than 150,000. The company, founded in 1998, has expanded beyond its core search engine-web browser-email core to artificial intelligence, phones, cloud storage, and a variety of other activities.
But in 2007, Google was the hot child of high tech, and was making its location decision for a new large regional server farm, or “data center…a building with powerful computers used to run a company’s services according to Google’s own perky video. “It’s where information is processed and made available. Data is stored, managed, and disseminated across these computers. And network infrastructure is installed to support web apps, databases, virtual machines, and more.”
Google rightly touts the efficiencies of reduced response times from having decentralized data centers, as well as the efficiencies of “multitenancy” in having locally centralized rent-a-space available at data centers rather than operating small, inefficient machines and software at each local business. Google is right, this is a good business to be in. They recognized this back in the mid-2000’s, and started building data centers all over the world. There are at least 12 in the US, with installations in Council Bluffs, Iowa; Douglas County, Georgia; Lenoir, North Carolina; Mayes County, Oklahoma; Montgomery County, Tennessee; The Dalles, Oregon; Berkeley County, South Carolina; Jackson County, Alabama; Clarksville, Tennessee; New Albany, Ohio; Pryor Creek, Oklahoma; and Henrico County, Virginia.
It’s the facility in Lenoir, a city in Caldwell County, North Carolina, that is of interest in my story. NC offered $260 million in “incentives,” taking the form of tax reductions, taxpayer-financed beefing up of electrical connections, and direct cash payment of taxpayer funds.
There were protests at the time from many quarters. But as David Barlow, then Mayor of Lenoir put it, “I truly believe that 80 percent of the people [of the town of Lenoir] are tickled to death, and they are getting mad” about the opposition and criticism.
Um. Well, yes. There is an old rule of politics: If you take money at gunpoint from A, and give it to B, B is more likely to vote for you. In this case, “A” is all the businesses, and taxpayers, of the state of North Carolina who are being forced to pay $260 million for a Google Data Center. “B” is the 100 people in Lenoir who get “new jobs” that pay an average of $48,000. Now, to be fair, that’s $48,000 per year. But we’re still talking about $150 million in total salaries benefit, and $260 million in costs over 30 years.
The town of Lenoir (B) is not paying the $260 million; That’s being paid by the rest of the state and all those chump businesses that located in NC for the right reasons, rather than engaging in site-selection extortion. It’s a little surprising that the good citizens of Lenoir only support the Data Center 80-20, given that they get all the benefits and pay only a tiny fraction of the costs.
If I’m right, then states should not be playing the incentives game at all, and should be trying to attract businesses for the “right reasons,” as I said in the previous paragraph. But what are those?
There is a frustrating “Catch-22” in public policy. It has been clear at least since Adam Smith wrote his manual of good public service, often called The Wealth of Nations. The book was not primarily a work of abstract economic theory, but was rather a guide for policymakers who cared about the public good. Smith noted that businesses would likely pursue profits through rent-seeking rather than competition, given the opportunity.
People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices. It is impossible indeed to prevent such meetings, by any law which either could be executed, or would be consistent with liberty and justice. But though the law cannot hinder people of the same trade from sometimes assembling together, it ought to do nothing to facilitate such assemblies; much less to render them necessary. (Emphasis added.)
Oh, but that “ought” is fraught, folks. Smith assumes that the goal of “the law” is the furtherance of competition, low prices, and high quality. Unfortunately, as the Public Choice school of political science was shown, there is no reason to believe that policymakers care about the public good unless the “public good” is defined as (1) reelection for legislators and (2) increased budget and power for bureaucrats.
The problem of achieving growth, at least in US states, which can rely on the American Constitution, financial system, and infrastructure, is well understood. There are three things that are — together — sufficient, to create substantial and persistent economic growth. Those three are the “right reasons” for businesses to make location decisions. Here they are:
1. Low taxes, little obtrusive regulation, and accommodating site selection
2. An educated workforce, with a substantial cadre of qualified graduate and undergraduate degree-holders
3. A market for housing that allows the building of a wide range of new rental and owner-occupied units, rather than NIMBYist pressures to limit growth
Since these three things together produce growth, isn’t it clear that this relatively simple policy program “should” be implemented by state legislatures? Adam Smith would have thought so, because he (charmingly) thought that the goals of legislators corresponded with the public good.
The problem is that it is very difficult for members of a legislature to claim credit — in the specific sense of “You got this because I did that” — for a good set of policies. The political scientist who made this argument most clearly was David Mayhew, who defined “credit claiming” this way:
[A]cting so as to generate a belief in a relevant political actor that one is personally responsible for causing the government, or some unit thereof, to do something that the actor considers desirable…The emphasis here is on individual accomplishment, and on the [legislator] as doer… [Much of legislative] life is a relentless search for opportunities to engage in [credit claiming].
It is not clear just what Adam Smith’s religious commitments were, but he was likely at least heavily influenced by the deist view of “God as watchmaker,” in which a system is set up optimally and then naturally performs well. An economic system that is set up optimally will likewise perform well, indefinitely. And Smith seemed to have thought that the legal system could be set up as an optimizing system too. But the temptation for elected officials to defect from the socially optimal set of rules, in ways that benefit them, the officials, is powerful.
And it’s not just selfishness, I should note. If a legislator actually runs on the socially optimal platform of low tax/education/housing, he will lose the election to a legislator who is willing to create opportunities for photo opportunities with BSOs. That’s “Bright Shiny Objects.” The way to get reelected is to promote the building of BSOs that would not exist if the legislator had not taken a specific action. As I have written about in this space several times, the use of discretionary incentives is the most common form of credit-claiming action for state officials. So even if a legislator doesn’t want to sell out for BSOs, he is obliged to do so or else become an “ex-legislator.” Selection by election ensures that BSOs win.
So let’s put all that together: Elected officials are no worse, but no better, than the rest of us. They respond to incentives. Given the choice between using the “right reasons” to attract new business and jobs, or using a very public pursuit of BSOs in a way that allows them to claim credit for the “recruitment,” they will go with the bad policy — particularized incentives and targeted recruitment — every time.
This is obviously “legal,” because legislators pass laws that make such particularized recruitment easy. Why would a rational, profit-seeking business — take Google, for example — not accept those kinds of substantial bribes in deciding where to locate? Profits that come from tax breaks and subsidies count just as much for the bottom line as profits that are earned by making better and cheaper products.
Still, back in 2007, I consented to be a named plaintiff in the lawsuit against NC, challenging the constitutionality of this public pimping. It violates, or so we said, “exclusive emoluments, public purpose, fair and equitable taxation, and uniformity of taxation provisions…[as well as] the law of land provision of the North Carolina Constitution.”
In English, that means that the clause of the NC Constitution that says “No person or set of persons is entitled to exclusive or separate emoluments or privileges from the community but in consideration of public services” actually means what the words appear to mean. As for “public services,” the discriminatory payments to Google, but not to other businesses that also create jobs, were not a contract for goods or labor but were instead a cash bonus to a private entity. “Uniformity of taxation” means, or should mean, that the state can’t say, “You: we like you. You don’t have to pay taxes. But you other people, watch yourselves. You have to pay extra taxes to support this other company that we favor.”
The case was dismissed, and then dismissed again on appeal, for “lack of standing.” The plaintiffs were all taxpayers in NC, but we had no actual say about how those payments taken from us were to be spent, even if the spending violated the Constitution of the state.
In the fifteen years since, Google has grown enormously. And fair enough, because Google provides a wide variety of useful services. The value proposition for fostering “local centralization” comes from creating data centers that gather many digital functions for businesses and private citizens, but that operate regionally so that there is fast communication and robust redundancy in the system. It’s a profitable business. It is not contingent on public subsidies or tax breaks the way “green energy” is; It’s a profitable business on its own.
But more and more states have entered contests for “buying jobs,” using the tax money of the businesses that already employ workers in the state. It’s a mook’s game. The problem is that the political logic is precisely designed to shake down mooks. Like us. I fought the Google, folks. But the Google won, and always will unless we cut back on the capacity of state legislators to buy votes using our money.
