Magical Stations: People, Petrol, and Policy

Is it possible that the wages of a person pumping fuel at the petrol station are not only disadvantageous to the employer, but have such a negligible effect that wage drag is minimal and competitiveness with competing jurisdictions with no equivalent workers is maintained?

“[A]n entire generation pumping gas, waiting tables.” – Tyler Durden, Fight Club

“It is a hazard for an untrained motorist to operate the pumps…” – testimony offered to the New Jersey legislative debate, Committee Session (25 November 1948)

“There are strong, eager men returning from the War and without work. We must use all our wiles to lure these men west.” – Gov. Douglas McKay of Oregon, Legislative Plenary Session (1951)

About a year ago, I was at Goldman’s 200 West offices meeting a friend who I’d met when we were master’s students at the University of Chicago. My friend suggested we take advantage of light mid-day traffic and visit her partner at Princeton. I’d made the trip many times on the PATH, but generally do not drive when I’m in the city, so this would be a new adventure.

We drove across to New Jersey, had a delightful lunch with her boyfriend, and then stopped for fuel. Her Bentley’s thirsty twin-turbo W12 engine pulled into the Shell station and shut down abruptly, as short-stroke W-configuration engines do. She looked over to me and said, like a proud coupon-clipper might at the grocer’s till, that she “fills up in New Jersey because it’s cheaper.”

Then there was a strange silence between us when we, as economists interested in empirical work, contemplated how this could possibly be true.

In New Jersey, it is illegal to pump one’s own fuel (only New Jersey and Oregon, two of America’s fifty states, maintain this policy - ostensibly as a safety measure, but practically as a source of employment for young men). Even if this fuel attendant’s wages were meager in the extreme, in the hypercompetitive and famously-efficient market for fossil fuels (where fractions of a penny are beaten out of the spread in less than a second by computerised trading mechanisms), the cost of petrol in New Jersey should not be lower.

As the young man refilled the leather-lined bulldog from Crewe, we discussed possible explanations. The difference in tax rates was not so extreme as to make up for the cost of full-time fuel-pumping employees, she observed. I noted that real estate is substantially cheaper in New Jersey, but that such attractive franchise financing is available for the sites (including everything from scaled incentive commercial site lending to sub-market-rate balloon mortgages) that it was hard to believe there was such a difference in operating costs net of financing.

The retail cost of fuel is fascinating.

Utah, which has miniscule refinery capacity and little oil extraction, consistently has the lowest retail fuel prices in America for both gasoline and diesel (this suggests the market is so efficient in fuel that transmission and transport prices for fossil fuel energy are flattened). Just behind Utah, consistently, is New Jersey, despite the financial drag created by its unnecessary pumping employees. New Jersey’s prices are consistently lower than adjacent New York, including areas outside New York City proper. Perhaps, one might hypothesise, the taxation of fuel in New York is so momentous that it not only meets, but exceeds, the cost of additional labour among fuel stations in New Jersey. This is quickly debunked, however, by departing from the trading price of the commodity itself and instead turning one’s attention to the labour market.

Station attendants in New Jersey are textbook overpaid workers. They are performing a task that the driver of the vehicle could perform herself (or himself) and, in all but the most extreme weather, might prefer to do herself (or himself). They are adding little, if any, value to the firm for which they toil. Their jobs exist as the result of a legislative flourish rather than of a mercantile necessity. If we assume, generously, that the pump attendants in New Jersey add five dollars per hour of value to their employers’ enterprises by helping senior citizens who would find it troublesome to pump their own fuel, by allowing a person in the minimart area to attend to the retail affairs, and by keeping the pump area in a janitorial sense. It is difficult to imagine the tasks assigned to such a worker, however diverse, fully justify (or compensate the employer for) the minimum wage of $7.25 per hour (a wage floor shared by New York and New Jersey). A busy station has enough revenue to dilute the effects of this (arbitrarily estimated) $2.50/hour gap between marginal wages and the attendant’s marginal product of labour. But dilution is not an answer to this problem; no matter how diluted the premium, or over how many tanks of fuel the overpayment of wages is amortised (if visualising the premium in accrual terms), the premium exists, whereas it does not exist in New York.

Interestingly, the full-service-jurisdiction-presents-savings dynamic is not true of the obvious comparative case: Seattle and Portland. Seattle and Portland are farther apart, but are similar in that neither is a refining or extraction centre. Despite differing state and municipal tax and employment policy (these differences are substantial: Oregon has no sales tax and requires fuel to be pumped by attendants; Washington has a significant sales tax and does not require fuel-pumping attendants), fuel prices are very similar between the two jurisdictions (as of this writing, on 6 May 2014, the median retail price in Seattle for standard-grade petrol is $3.64 per gallon and $3.63 per gallon in Portland). It is tempting, of course, to say that perhaps the sales tax in Washington and the employment cost of fuel attendants in Oregon are similar, but the sales tax in Washington is an order of magnitude smaller than the marginal wages of the attendant in Oregon. Commercial real estate costs are similar between major municipalities the two jurisdictions and the effective interest rate for commercial first-tier retail site financing is likely similar.

On our drive back into the city, weaving her coupe through reverse-commute traffic, we continued discussing the question, dismantling each others’ hypotheses. It remains unclear how New Jersey can maintain such consistently lower retail fuel prices than neighbouring New York. The only explanation either of us could support remains the extramarket one I believe is true one year later: that New Jersey substantially (more substantially than is generally thought) subsidises fuel station operators, possibly through zoning-specific real estate tax credits or other relatively opaque subsidy schemes, to allow them to be competitive with New York, figuring that increased business to New Jersey’s petrol stations not only creates velocity in that retail market, but also steers dollars from New York toward other New Jersey businesses. Without a subsidy scheme at least equal to the difference between the productivity of New Jersey fuel pump attendants and their marginal wages, it is very difficult to explain New Jersey’s competitive retail fuel prices.

The search for a more detailed explanation, however, goes on.