JLARC and the Virginia Cloud

Facebook data center in Henrico

Some years ago, the General Assembly made considerable use of the time between sessions. There were special study commissions that met fairly frequently, as well as meetings of subcommittees of standing committees. For various reasons, that does not happen much now. As a result, the legislature has struggle with tough issues, with little time for research and reflection, during the crowded regular sessions.

More and more, the Joint Legislative Audit and Review Commission (JLARC) is filling the void, providing the legislature with analyses and background on a number of thorny issues. This is a positive development.

Originally, the primary function of JLARC was to conduct regular, thorough analyses of agency operations. Gradually, that function has evolved to consist of (i) ongoing oversight over VITA, VRS, economic development incentives, Virginia529 (college savings plan), and Cardinal (the state’s new accounting system); (ii) several annual reports on state spending; and (iii) specific topics referred to it by the legislature or taken up on its own initiative.

This year the agency has one of its heaviest study loads ever. In addition to the ongoing oversight and state spending studies, the workload includes studies on community services board funding, implementation of STEP-VA (reform of the state’s behavioral health system),  the Office of the State Inspector General, VITA’s new infrastructure, Office of the Attorney General, gaming in the Commonwealth, Medicaid expansion, Workers’ compensation, and the Department of Game and Inland Fisheries.

The JLARC staff is professional and well respected, well versed in quantitative and policy analysis. The Commission itself consists of senior members of the legislature and is chaired this year by Sen. Thomas Norment, who is the Senate Majority Leader and co-chair of the Senate Finance Committee.

Earlier this week, the staff presented another report in its series on the evaluation of economic development incentives. The report covered 11 incentives intended to promote data center and manufacturing growth in the state. This post will focus on the main incentive: the data center retail sales and uses tax exemption.

The incentive exempts an eligible data center from having to pay sales tax on the purchase of computer equipment and items used to operate that equipment. To be eligible for the incentive, a data center must make a new capital investment of at least $150 million and create at least 50 new jobs in the locality in which it is located.  For enterprise zones or areas with high unemployment, the jobs threshold is lowered to 25.

The data center exemption is the state’s largest economic development incentive. JLARC calculates that it represented more than one-fifth of the state’s total incentive spending between F Y 2010 and FY 2017. It is estimated that the data centers saved $417 million in sales tax expense over that time period. Conversely, that translates into $417 million in tax revenue given up by the state. Of course, it could be argued that, without the incentive, the data centers may have located in another state, as they did in North Carolina before Virginia broadened its exemption. In such cases, the state also would not have collected the sales tax revenue.

Virginia has become the nation’s leader in the data center market. There are 159 data centers benefiting from the exemption. The vast bulk of the state’s data center activity (69%) is in Northern Virginia. It is well known that an estimated 70% of the international internet traffic flows through the telecommunications infrastructure in Loudoun County. The Richmond region is second with 15% of the activity. Surprisingly, Southside Virginia is close behind Richmond, with 10%.  The Southside activity is the result of a Microsoft data center located in Mecklenburg County, near the town of Boydton. That center opened in 2010 and has expanded nearly every year since.

According to JLARC, the sales tax exemption has “been a key factor in the [data center] industry’s growth” and, therefore, appears “reasonable” and “relatively effective.” JLARC did not define what constitutes being effective, unless it means succeeding in attracting data centers to the state. The Virginia Economic Development Partnership has identified data centers as one of its targeted industries.

If one defines “effectiveness” in terms of improving the state’s economy, the picture is more nuanced. Even JLARC rates the incentive as having a “moderate” economic benefit to the state and showing a “moderate” return in revenue. JLARC data shows the following average annual net impact to the state over the period FY 2010-FY 2017:

  • Private employment: 7,665 jobs;
  •  Virginia GDP: $1.3 billion;
  •  Personal income:  $724.9 million.

However, as the report points out, most of the economic impact “occurs during construction of the data center and from initial and ongoing expenditures for equipment, rather than the daily operation of the data center.” Therefore, unless the recent round of construction of data centers continues, that economic benefit will lessen.

The impact on state revenue is even less encouraging. JLARC calculated that, for the period of FY 2010-FY 2017, the average annual total revenue generated for the state by the data centers taking advantage of the incentive was $37.7 million, whereas the loss of state revenue resulting from the incentive was $52.2 million, a return of $0.72 for every $1 spent. In other words, the incentive does not pay for itself, although that is often the justification given for providing an incentive. JLARC summarizes by saying, “This exemption is much more expensive in terms of direct job creation and capital investment than most other incentives.”

But, that is not the whole story. Because JLARC’s charge is to evaluate the effect of economic development incentives at the state level, its data and analyses do not include the effects on localities, except in passing. Although the localities in which the data centers are located lose out on their share of sales tax revenue due to the exemption, they more than make up for it in revenue from real property taxes and the machinery and tools tax. According to JLARC, Loudoun County collected more than $154 million in property taxes from computer equipment in FY 2018 and Prince William County received more than $25 million in FY 2017.  No data is included for real property tax revenue.

“Localities with data centers reported data centers are important sources of tax revenue, and they do not require substantial local government services,” concludes JLARC. In summary, the data center exemption is a costly incentive with which the state subsidizes highly sought-after facilities that are revenue producers for localities, primarily in Northern Virginia. It could be argued that the subsidy is justified because, by benefiting localities, it benefits the state as a whole, and, without it, data centers may locate in other states.

JLARC made three recommendations:

  1. Reduce or remove the minimum job creation requirement for the sales and use tax exemption for data centers locating in a distressed area or an enterprise zone. This action could result in more data centers, especially smaller ones, being located in rural areas of the state.
  2. Convene a work group to examine actions that could be taken to maintain the state’s competitive position to attract data centers and examine whether the opportunity exists to reduce the level of the exemption without adversely affecting industry growth.
  3. Better track data center investments to improve estimates of fiscal impact and economic benefits.
There are currently no comments highlighted.

10 responses to “JLARC and the Virginia Cloud

  1. Dick – your posts are even MORE wonkish than Steve’s!!!

    But they are excellent – and non-partisan … apolitical = and feel
    as if both pros and cons are presented.

    They also show that if one looks comprehensively at some of these things that such issues (like ED) are not so black and white… AND that Government – is much more than a bunch of citizen legislators blindly taking votes in the GA – some serious in-house analysis is going on – and I presume it is objective and balanced to yield the unvarnished truth (that may be going too far, (I’ve heard that sometime JLARC in the past has been accused of reporting with a somewhat political focus).

    Don’t want to tell you how to write but have you considered some sort of Sizemore’s Sizzling Summary for those of us who need help in summarizing the main points? 😉

    Finally – a request. I wonder if (in the future) you might also be encouraged to write about another State Agency called the Auditor of Public Accounts that produce (among other things) the Local Government Comparative Reports
    (All Virginia counties, cities, towns with a population of 3,500 or more, and towns operating a separate school division are required to submit this data to the Auditor of Public Accounts annually. The data presented represents the local government operations for the general government and enterprise activities.).

    I see these reports as fertile ground to compare localities on how they allocate and prioritize their revenues and spending.

    But there are other reports also and I’m not quite sure the distinctions between APA and JLARC sometimes although both are clear assets to Virginia Government.

    At any rate – thank you for your tomes… I can tell that you do put a lot of effort into them but the knowledge you have and share is invaluable.

    • Larry, thanks for the suggestions. This one was a little long because I wanted to provide a little background on JLARC and to lament some on the GA’s decline in educating itself betweeen sessions. I will put the APA on my list for future posts.

  2. Dick, we’re going to have to start coordinating. 🙂 I’ve been busy with pressing personal business this week, and held this morning open to read and post on the JLARC report. Still going to read it. Whether or not this emergence of JLARC as a counterbalancing force is highly positive remains to be seen. It is certainly getting far more active, but if/when the General Assembly flips it may become a major force with a different point of view entirely.

    • re: ” .. it may become a major force with a different point of view entirely.”

      that makes it sound like JLARC might be affected by partisan influences….and reports might be different on the same subject depending on whose directing….

      I’ve actually heard that before but was disappointed to hear it because it undermines one’s views that such reports are objective and credible.

  3. Great minds…. I would like to know if you agree with my interpretations.

  4. Ok – I’ve started reading. Plenty for another report….

    Larry, hey, language matters. It sends a message. This report starts by treating any tax policy preference as “spending.” Spending on this and spending on that. The bottom line assumption behind that is the money belongs to the government, and by allowing the taxpayer to keep it that’s “spending.” And this is when JLARC is managed by a majority-GOP panel. You don’t think the view might tilt a bit when the other party is setting the study agenda?

    The reports are very credible, the data reliable. I’m a JLARC fan (much more fun than plowing through APA data.) But it can have a point of view, and we’re free to dispute the conclusions or recommendations.

    • I don’t think the substance of the reports would be affected by the partisan makeup of the Commission. However, the choice of study topics would probably be influenced.

    • Steve, I’m familiar with the “logic” but govt finances are all about priorities as there is never enough money to buy everything that everyone wants.

      I’m a JLARC fan as long as they keep their partisan impulses under control – both sides.

      The argument about “spending” and “investment” is partisan. JLARC should be dealing in facts and resisting those who would push them beyond that.

      To this point, I’ve always considered JLARC and APC as related govt truth-tellers. (I thought that about Weldon Cooper also).

      Once they stray from that mission – they’ve forever damaged their credibility and reputation and just become yet another partisan arm of some group.

      Too bad. Not the end of the world. I just would no longer consider them a gold standard and instead one voice that needs to be compared and contrasted to other voices.

      Local and State Govt collect taxes – and they spend that money on things most folks think are legitimate. There are always arguments about how much to collect and what to spend it on – but no one, even those on the right – (except for a few idiots) are going to advocate that the govt stop collecting taxes because when they do – it takes money out of the pockets of citizens or some such partisan idiocy.

  5. “1. Reduce or remove the minimum job creation requirement for the sales and use tax exemption for data centers locating in a distressed area or an enterprise zone. This action could result in more data centers, especially smaller ones, being located in rural areas of the state.”

    And the rural datacenters would get high speed network connections via what means?

  6. Pingback: Pollution Control Tax Break Not An Incentive - Bacon's Rebellion

Leave a Reply