Richard Drew, AP
In this March 29, 2018, file photo, the logo for Facebook appears on screens at the Nasdaq MarketSite in New York's Times Square. The company will construct a data center in Eagle Mountain, Utah.

SALT LAKE CITY — A little more than 10 years ago, Google decided to pit North and South Carolina against each other in a competition to win a large, new data center. Only after getting promises of big tax incentives from each did the company reveal that (surprise!) it was going to build two centers after all, one in each state.

Stop me if you’ve heard this before.

Facebook’s decision to build data centers in both New Mexico and Utah, two states it had pitted against each other in 2016, may have been far enough apart to make it a stretch to say this was part of a larger strategy all along. But no one should be too surprised that Facebook came back to the Beehive State — or that it played us for a lot of money.

Mega high-tech companies look for certain things when building data centers. The first is safety. Areas prone to natural disasters — hurricanes, tornadoes, floods, earthquakes — are not ideal.

Good news, there. A Wallethub.com study released this week showed Utah is the fourth overall safest state in the union, also ranking fourth best in terms of loss claims per capita due to climate disasters.

Companies also prefer dry climates and low electricity costs. Utah does well there, too, ranking 11th cheapest in the country for power, according to the site chooseenergy.com.

It was a cinch. Facebook was coming back. And it was going to demand tax breaks because, well, that’s what everyone else is willing to give.

Only a politician or an economic development official can make giving millions to one of the world’s richest companies — one that earned $12 billion in the first quarter of this year alone — sound like a great deal.

So, why write about this again now, after the Facebook deal has been approved by each of the taxing entities affected and construction is set to begin? Because once the papers are signed and the shovels start turning, people tend to forget about accountability. But that’s when accountability should begin.

The initial tax benefit will be $150 million over 20 years. Considering the center is expected to employ up to 50 people, this equates to about $3 million per job.

But of course it isn’t that simple. The company has agreed to make infrastructure improvements, also estimated at $150 million, which is expected to make further development in that area possible. Others have touted a data center’s ability to attract other businesses and growth. Of course, lots of temporary construction jobs will be created during construction.

But when do we ever come back years later and commission independent studies to see if the promises come true? When do we weigh the tax breaks against the gains over time and see if taxpayers end up in the black?

The Utah Foundation just released a study on sales taxes. Among its findings is that Utah has granted 91 separate exemptions to the tax, which cost the state hundreds of millions of dollars per year.

Anyone who has been to the Legislature knows how attractive each of those exemptions sounds when it is proposed. And yes, state lawmakers have even created some especially for data centers. But no one ever mentions what one of these will add to the total.

The information at the beginning of this column came from a study two years ago by a group called Good Jobs First. That study agrees with the premise that data centers can be good for a local economy, but not if local governments give away all the tax benefits.

The study recommends capping tax subsidies at $50,000 per job, asking governments to consider whether any private business would agree to a deal that equates to millions of dollars in investment per job.

32 comments on this story

Second, it recommends transparency. Don’t keep the name of the company secret while local governments are debating tax subsidies. Let people know all the details.

I would add one more. The state should set up a public database tracking all such subsidies. While it may be impossible to accurately estimate the gains brought in by one single project, a complete database at least would let people know how much of their money is being given to private businesses, and to which ones.

If all of this keeps mega companies from coming here in the future, it may be instructive to ask why.