Electricity Policy

“Do Renewable Portfolio Standards Deliver?” by Michael Greenstone and Ishan Nath. May 2019. SSRN #3374942.

States have responded to concerns about greenhouse gas emissions from coal and natural gas in electricity generation by enacting laws that mandate the use of renewable generation sources. Those who argue such mandates have low or no net cost emphasize estimates of the total costs of various generation technologies divided by their potential output over their operating lifetime, the so-called “levelized” cost of energy estimates (LCOE). By this metric, large-scale centralized solar generation in the deserts of the American southwest and large-scale onshore wind generation both have costs that are competitive with new natural gas generation.

However, even if the lifetime average costs of wind and solar generation are the same as coal or natural gas, the equivalence needs to be qualified. The first problem is that renewable generation sources are intermittent. The sun does not shine at night and the wind blows more at night than in the daytime. According to the authors, utility-scale solar generation plants’ annual output averages only 25% of their potential output. Wind plants’ output is only 34% of potential. In contrast, the output of natural gas combined-cycle plants that “always” operate is about 85% of potential. Thus, a comparison of LCOEs between intermittent renewable sources and “baseload” conventional technologies is very misleading with respect to total system costs because it does not account for the additional cost necessary to supply electricity when the renewable resources are not operating.

Until cost-competitive green energy that is dispatchable is available, renewable sources of electricity require backup natural gas generation whose output can be varied (sometimes quickly). The fixed and variable costs of the backup must be paid by someone. These hidden costs need to be considered in any calculation of “cost competitiveness.”

The second problem is that renewable generation plants are frequently located far from population centers, which increases transmission costs relative to those of fossil fuel plants. According to the authors, a literature review of transmission cost estimates for wind power by the Lawrence Berkeley National Laboratory finds a median estimate of about $300 per kilowatt, or about 15% of overall wind capital costs. This adds approximately 1.5¢ per kilowatt hour to the levelized cost of generation for wind. An analysis by the Edison Electric Institute in 2011 found that 65% of a representative sample of all planned transmission investments in the United States over a 10-year period ($40 billion) were to link renewable generation with the existing transmission system. These transmission costs are part of the total cost of renewable energy.

This paper attempts to calculate the total costs of renewable mandates on electricity prices by comparing states that did and did not adopt Renewable Portfolio Standard (RPS) policies, using the most comprehensive panel data set ever compiled on program characteristics and key outcomes from 1990 to 2015. Electricity prices increase substantially after RPS adoption. The authors’ estimates indicate that in the seventh year after passage, average retail electricity prices are 1.3¢ per kWh (11%) higher, totaling about $30 billion in the RPS states. And 12 years later they are 2.0¢ (17%) higher.

When the emission-reduction estimates are combined with the estimated effect on average retail electricity prices, the cost per metric ton of greenhouse gas abated exceeds $115 in all specifications and can range up to $530. That is several times larger than conventional estimates of the social cost of carbon.

Housing Policy

“The Effect of New Market-Rate Housing Construction on the Low-Income Housing Market,” by Evan Mast. July 2019. SSRN #3426103.

What policies increase the availability of affordable housing? One possibility adopted in New Jersey, Massachusetts, and New York City is to mandate that new housing developments have a small percentage of units set aside and priced for low- and moderate-income households. Another is to reduce policy constraints on new construction and allow the effects of the increased new supply to “filter down” to the vacated existing units whose owners have to reduce price to maintain occupancy.

This paper analyzes these policies by using newly available data collected from numerous private and public record sources such as U.S. Postal Service change-of-address forms, county assessor records, magazine subscriptions, and phonebooks. Each address is accompanied by an estimated date of arrival and some limited demographics (age, gender) on each individual. The data consist of 52,432 individuals in 686 market-rate multifamily buildings constructed since 2009. The buildings are relatively evenly distributed across cities, with Seattle, New York City, and Chicago having the most (over 80 each) and Philadelphia and Boston the least (under 20).

The data allow the construction of “migration chains” as people change dwellings. The data strongly suggest that a short series of moves connects new construction and low-income areas, meaning that as new construction expands housing supply, existing housing becomes available for lower-income renters and buyers. One hundred new market-rate units create vacancies in 70.2 units in below-median income tracts, 39.6 in bottom-quintile income areas, and 45.3 in areas that are below median income and in the top quintile of rent burden. Inclusionary set-aside requirements, which are in the range of 5–20% of the new units constructed, are much fewer than the 70% of unit vacancies created by market filtering. Market filtering would appear to be the superior policy for expanded low-income housing.

Health Care Policy

“The Opportunities and Limitations of Monopsony Power in Healthcare: Evidence from the United States and Canada,” by Jillian Chown, David Dranove, Craig Garthwaite, and Jordan Keener. July 2019. NBER #26122.

Medicare for All”—opening the Medicare program to people of all ages, either as a competitor to private health coverage or in place of it—has been embraced by some Democratic candidates for president. The economic reasoning supporting the proposal is a claim that a single buyer (a monopsonist) would reduce the prices paid for both health care labor and drugs. Canada has a single-buyer system and is often held up as an example to be emulated by the United States.

Many presume that Medicare for All in the United States would result in Canada-like health care wages and pharmaceutical prices. Highly educated Canadian health care workers earn 26% less than their American counterparts. But all skilled Canadians earn 22% less than Americans with similar credentials, so only 4 percentage points of the wage difference between Canadian and U.S. health care workers can be plausibly attributed to the monopsony power of the government. The Canadian government does not use its buying power to lower the wages of health care workers very much relative to other skilled workers because doing so would decrease the supply of health care workers who would turn to other skilled occupations if their wages were suppressed through policy.

Pharmaceutical prices in Canada are 54% less than U.S. prices, whereas overall Canadian prices are only about 4% lower than U.S. prices. So, Canada is exercising monopsony power in the pharmaceutical market. It can do this because it is a small purchaser in the context of the world market and its low prices will not reduce pharmaceutical supply to Canada. The United States is not a small purchaser and could not reduce pharmaceutical reimbursement 54% without having supply consequences.

These findings should temper the expectations of anyone who believes that Medicare for All would reduce U.S. health spending to Canadian levels.