Beware the Free Lunch: A Critique of the Proposal to Privatize Pennsylvania’s Wine and Liquor Sales
Over the past two years, advocates for the privatization of Pennsylvania’s state wine and spirits stores have asserted repeatedly that privatization would benefit the commonwealth economically, including by delivering a large upfront fee often estimated at $1 billion to $2 billion. Recognizing the lack of solid information to support these claims, the Corbett Administration last year commissioned a study by Public Financial Management Group Inc. (PFM) of the state revenue, tax, and consumer implications of a possible privatization of Pennsylvania’s state wine and spirits distribution system. The present report critiques the PFM study, Liquor Privatization Analysis, building on testimony delivered to the House Liquor Control Committee on December 1, 2011.
We document several major flaws in the PFM study.
- Starting with the simplest error, in the first year of its projections PFM’s estimate of the income of the public wine and spirits system was off by more than 50%. Actual data show that the public system generated $104 million in 2010-11, 54% more than the $67 million projected by PFM.
- In a second step in our analysis, we compared the actual income the state received from the state’s wine and spirits stores in FY 2010-11 with what the state would have received using the privatization scenario recommended by PFM. The result: Pennsylvania would have lost $96 million if the wine and spirits stores were private. Lower collection of sales taxes would have resulted in an additional $19-$20 million loss in 2010-11. Privatization, therefore, would have resulted in an estimated $113 million less revenue for the Commonwealth in FY 2010-11 than the current system.
- Third, we used the actual price paid for retail liquor licenses in West Virginia in 2010 as a test of the PFM projections for upfront license fees. Adjusting for differences in consumption and population, the actual amount West Virginia received for auctioning off retail liquor licenses projects to an amount for Pennsylvania licenses that is 41% below what PFM projects. While there are significant differences between the two states, more of these are likely to lower Pennsylvania auction fees further than the reverse.
Identifying other flaws in the PFM study requires probing more deeply the plausibility and internal consistency of the assumptions built into PFM’s economic models. These assumptions were shaped by an explicit charge that PFM received from the state to (a) assure that the state achieves revenue neutrality—the same amount of money annually from the wine and spirits industry as under the current system (e.g., through taxes, annual license fees, or transfers to the General Fund), while (a) maximizing the upfront sale value of liquor distribution franchises.
As part of PFM’s effort to achieve revenue neutrality, the consulting firm builds into its recommended privatization option
- annual retail and wholesale license fees that are at least four times higher than neighboring states, including West Virginia’s; and
- per gallon alcohol taxes that are six to 21 times higher than those of neighboring states on wine, and up to five times higher on spirits.
Both high annual license fees and alcohol taxes will tend to drive down Pennsylvania sales, and reduce profitability, which would make wholesalers and retailers less willing to pay a large sum up front for liquor licenses. To remedy this problem, PFM assumes in its model that both wholesalers and retailers would have operating expense that are low by industry standards which would allow them to profit with low margins (charging customers only a bit more for wine and spirits than distributors pay for the product themselves). The combination of low margins and low operating costs makes it possible, on paper, for companies to be profitable and for prices of Pennsylvania wine and spirits to remain competitive. The body of the present report, however, compares the PFM operating expense and margin assumptions with actual data and shows that these assumptions are unrealistic.
The underlying problem is the incompatibility of the two economic objectives that the commonwealth asked PFM to achieve in its projections: a big upfront payday from auction revenues and revenue neutrality on an annual basis. While high annual license fees, high sales taxes, and low “margins” all help achieve high revenues each year for the state, the same three assumptions make holding licenses less profitable and are thus incompatible with a large revenue windfall from the auction of wholesale and retail licenses. Pennsylvania cannot achieve both a big upfront payday and revenue neutrality if the sale of wine and liquor rights goes forward. As the saying goes, there is no such thing as a free lunch.
The PFM study is likely on target when it implies that rural consumers would see higher prices and less product choice as a result of privatization.
This critique of the PFM study does not address the impact of privatization on excessive drinking and associated public health problems and traffic fatalities. Those public health issues are addressed in an earlier policy brief and in legislative testimony delivered by the Keystone Research Center.
As of this writing, the PFM privatization scenarios are no longer the focus of legislative discussion about the state’s wine and spirits stores — and privatization itself may not be taken up by the legislature this session. Nevertheless, it is important to document the flaws in the PFM study in order to appreciate that there is no free lunch when Pennsylvania privatizes state assets, including her wine and spirits stores. Legislators should not support wine and spirits privatization based on the belief that it is a revenue winner for the commonwealth, because that notion is false, especially beyond the immediate election cycle.
Roland Zullo, Ph.D., “Beware the Free Lunch: An Analysis of Public Finance Management’s Liquor Privatization Analysis,” Testimony to the House Liquor Control Committee, December 1, 2011.