7. Conclusion: What Do We Know and What Should We Do?

There’s no better motivation for this than the Covid-19 pandemic. Were the lockdowns instituted in most states starting in March of 2020 good public policy? Answering this requires knowledge of their causal impacts, as well as making philosophical judgments about tradeoffs like the value of a statistical life. The pandemic clearly illustrates the strong social need for people who understand how to estimate the impacts of policy, and who also understand techniques for rational decision-making.

https://​founders.​archives.​gov/​documents/​Franklin/​01-19-02-0200.

See Boardman et al. (2017) and Fugiutt and Wilcox (1999).

Other methods of economic analysis include Economic Impact Analysis (EIA) and Fiscal Impact Analysis (FIA), which are often mistakenly described as CBAs. One of the goals is to section is to describe what CBA is, so a reader will be able to recognize when an analysis that is described as a CBA is in fact something else. The focus in CBA is on human welfare broadly conceived, while EIA and FIA are narrower and focus on specific impacts. For example, FIA may focus on the impact of some policy or program on the state government’s budget, while EIA may focus on GDP impacts. Meanwhile, CBA recognizes that social welfare can go up, even as state budgets and GDP go down. Examples of EIA and FIA, respectively, include Chaudhuri and Zieff (2015) and Culhane et al. (2002).

One of the Questions for Review at the end of this chapter discusses a CBA of reducing elementary school class size. The debate between Krueger (2003) and Hanushek (2003) highlights econometric considerations with the plug-in method, as well as the importance of Step 4, Impact Estimation, more generally.

The details behind this calculation are 48 kWh \(\times \) 12  months \(\times \) 14.6 cents = $84 annual electricity cost savings. For natural gas, the calculation is 1.5 therms \(\times \) 12  months \(\times \) $1.22, = $21.96. Adding these together, $22 + $84 = $106, which is the value of the annual energy savings.

It is a subtle point, but the reason monetizing with market prices is appealing is because neoclassical perfect competition theory teaches us that price equals marginal cost. However perfect competition is a theoretical condition not always met in the real world, and there are times when the use of market prices is not appropriate. In these cases, analysts have to be creative in calculating a so-called shadow price, which is simply the true social value of an impact.

Their Table 5 indicated the low- and high-end figure for carbon dioxide reduction due to fall in natural gas use was $0.83 and $10.12, respectively. Thus the marginal damage figures are backed out as follows: $0.83/0.108 = $7.68, and $10.12/0.108 = $93.7. Subsequently, I verified with the authors this was the marginal damage figure they used.

NPV can be computed using R, in a spreadsheet, or with and pencil and paper using the following handy formula: \( -675 + 190*[ (1/.05) - (1/.05)/(1.05^{10}) ] \). There is an intuition behind this term in brackets. The term 1/r is the present value of a dollar received every year forever (an annuity that pays out forever is called a perpetuity), which is $20 when r = 0.05. The second term in brackets is the present value of $20 ten years from now, which is 12.28. So the term in brackets can be thought of as the present value of a $1 perpetuity that is taken away in ten years. This is $20 minus $12.28 or $7.72. Given there are $190 in benefits every year for ten years, we multiply $190 by 7.72 to find $1,467, the present value of benefits. From this we subtract $675 which is the initial upfront costs to find a present value of 1,467 minus 675 or $792.

How does this NPV estimate compare with the decision criteria presented in the JK analysis? Jacobsen and Kotchen presented three criteria, the first of which is a private payback period, which is calculated as the upfront costs of $675 divided by the annual savings of $106 which comes to 6.37  years. This is the amount of time it would take a homeowner to recover their investment in the thicker windows. This criterion assumes a zero discount rate and does not account for impacts on third-parties, but is easy to interpret. The second criterion could be called a global social payback period, which is the upfront costs of $675 divided by $190, the sum of private and social benefits, and comes out to be 3.5  years. Third, Jacobsen and Kotchen recognize that, “...one might argue that the benefits associated with a lower CO2 emissions should not be considered...as they are likely to occur for the most part outside the policy jurisdiction” (p. 47). Excluding CO2 reduction benefits reduce the value of emissions reductions from $84 to $22, and what could be called a national social payback period rises to 5.3  years ($675 divided by $128, where $128 is the sum of $106 and $22.) A decision maker (homeowner, policy maker) would have to somehow determine a cutoff value for the payback period to make a decision regarding low-E windows.

Florida’s energy code gives the builder flexibility about how to meet the energy use requirements specified in the home. If there is a design change that enables the builder to comply with the code more cheaply than by using low-E windows, the builder could select that design feature instead. This means that the $675 figure might overstate the actual cost of compliance—though I still refer to $675 as the low-end estimate.

One further consideration is worth highlighting. The analysis described above was for a representative home. If all homes in the study area are basically the same, we could simply multiply the NPV for a single home, which is what we calculated above, by the number of homes. A more careful analysis would have to account for the fact that homes differ. It turns out, many homes in Florida do not use natural gas at all; evidence from the ACS suggests only about 25% of recently constructed homes use natural gas.