It's a familiar situation: business and government officials are promoting a new economic development deal which naturally includes subsidies for the investor. They may be touting a consultant's study touting massive ripple effects and fantastic taxpayer return on investment. Should you believe them?
Of course not. Consultants, whether they technically are working for the company or the government, don't get paid to say deals don't make sense. Or at least they don't if they want future contracts with the company or city/state/province/country paying them.
What's a taxpayer to do? First, you need to know that uncontrolled subsidy competition like we have here in the United States is going to systematically provide higher subsidies than will a rules-based system like that of the European Union. The EU's state aid rules guarantee transparency, restrict location subsidies to the poorest regions, and cap the amount of subsidy a project can receive based on two principles: 1) The richer the region, the lower the maximum subsidy allowed (with 0 as the maximum in many regions); and 2) For large projects, the maximum subsidy is reduced below the region's normal percentage cap (determined under the first principle) with a 50% cut in the cap for projects over 50 million euros and a 66% cut for the subsidy on the amount invested over 100 million euros. So you need to remind elected officials and economic developers of the need for transparency and real limits on subsidy use.
Realistically, though, we are a long way away politically from having rules like the EU's. Right now, we're making progress on transparency but not so much on substantive rules, not even anti-piracy agreements. What do we do to evaluate investment incentives now?
The goal in bargaining is to get the most benefits for the least cost. How do we know if the cost is high or low? This is where the Good Jobs First Megadeals database comes in. We can use it to compare the cost with similar deals made elsewhere in the United States. You can use Subsidy Tracker if you want to look at smaller deals, but for my purposes in this post I can't deal with a quarter of a million projects, so I am sticking with Megadeals.
Projects vary widely in size, so to compare different subsidies we need metrics, or standardized ways of measuring them. Knowing the gross cost of a subsidy is not enough (and technically, using the present value of the subsidy rather than its gross cost would be better, but would not solve this problem). As I wrote in a report for the North Carolina Budget and Tax Center, $1 million would be a large subsidy for a call center, which needs little more than computers and phone lines. But $1 million would be miniscule for an automobile assembly plant, which can cost as much as $1 billion to build.
The good news is that we don't have to reinvent the wheel to obtain these metrics. Two good ones already exist, cost per job (subsidy divided by the number of jobs) and what the European Union call "aid intensity" (subsidy as a percentage of investment). The EU's rules set limits as discussed above in terms of aid intensity, but using both metrics together is even more illuminating.
Better news still is that the Megadeals database already contains the data needed to calculate both cost per job and aid intensity. Since you can download Megadeals in spreadsheet form at the link above, I did just that and then added two columns (cost per job and aid intensity) to create a spreadsheet with these metrics. Based on an informal conversation we had in March, Christopher Lau, a Senior Advisor in the Ontario Ministry of Economic Development Trade and Employment, in an effort outside of his official capacity, added a chart grouping all the automotive investments together. This is on the larger of the two spreadsheets at the link below:
https://sites.google.com/site/middleclasspoliticaleconomist/megadeals-with-metrics
Christopher's chart (see below) lists all the projects from largest to smallest aid intensity, so it's a good place to begin to see the range for an auto plant. We can see from the list that the aid intensities are pretty tightly bunched from 15% to 41%, with no gap greater than 4 percentage points within that range. We can presume that anything above 41% is overpriced, especially the Gestamp project, since a stamping plant is one of parts facilities you would hope to have locate near an assembly plant, which means you would normally subsidize it less than an assembly plant.
Even within the 15-41% range, we shouldn't conclude all are reasonably priced given the situation states and cities are currently caught in. We would want to factor in when the facility was built, focusing on the most recent experience when discussing a potential new project. We would want to know the location, the size of the project and the number of jobs created. And we need to remember with regard to jobs that if we build a new assembly plant, its output will reduce the sales and jobs at existing facilities, so the country's net job gain will be much less than the total employment at the new plant.
This post illustrates how we can use the Megadeals database to help evaluate the cost of a proposed subsidy deal. In my next post, I want to reiterate all the best practices that should be attached to any deal before we can say that it is a relatively good one under the United States' current Wild West subsidy system.
Source: Calculated from the Good Jobs First Megadeals spreadsheet.
Note: Christopher Lau's analysis in this article is his own and does necessarily reflect the views and opinions of the Ministry in which he is employed.
Cross-posted at Angry Bear