Policy-Driven Costs and Savings are Not “Good” or “Bad”

There is not just one correct way to calculate the financial cost or savings of a set of policies, and even the same type of cost (or savings) may be seen either in a positive or negative light.

Policies can increase or decrease spending on:

All of these changes in spending are calculated in the Energy Policy Simulator.

However, the simulator can’t tell you whether a change is “good” or not. For example, a policy that improves the lifetime reliability of industrial equipment would reduce capital spending (because factories need to replace their equipment less often). This “saves” the business money, which sounds good. But it reduces economic activity (reduces GDP), by reducing the sales of the companies that manufacture the equipment, and the companies that make the materials (like steel) of which the equipment is made.

GDP is itself not a clear “good.” This is because not all spending is equal. Spending on food, entertainment, and leisure travel are often considered good, but higher spending on disaster recovery or medical expenses is often considered bad. A policy that worsens air quality and sickens people, so they must visit the doctor more often and buy expensive medicines, will tend to increase GDP (even after accounting for lost workdays). This is especially true if the policy primarily sickens retirees, who are not working. But no reasonable person would argue that making retirees sicker is a good policy outcome, even if it boosts GDP.

Similarly, spending more on labor can be seen as a positive thing (job creation), and spending less on labor can be seen as a positive thing (automation increases domestic businesses’ productivity and competitiveness).

A wise policymaker thinks not only about whether a policy increases or decreases certain types of costs, but what underlying dynamic drives those changes, and whether that dynamic makes the country (and the world) a better or a worse place to live.

Cost Metrics in the Energy Policy Simulator

The Energy Policy Simulator includes two main cost metrics:

First-Order Cash Flows and Higher-Order Economic Effects

The Energy Policy Simulator only calculates the “first-order” financial effects of a policy package: Who gives how much more (or less) money to whom? It does not consider what the recipient does with increased money, or what the spender does with savings from reduced expenses. But how the money is used can have a large impact on the policy’s outcome. A carbon price that raises government revenue will go farther if the government wisely spends the money (for example, on support for research and development, which can accelerate technological progress, or improving public transit systems, as public transit generates economic value several times greater than its costs). If the revenue is spent unwisely, the policy will not do nearly so much good.

Our tool is an “Energy Policy Simulator,” not an “Everything in the Economy Simulator.” We consider decisions about how government spends additional revenue to be outside the scope of the simulator. (The same is true for other tracked cash flow entities, like “labor and consumers” or “non-energy industries.”)

Therefore, the graphs that show “Financial: Cash Flow Changes” should not be interpreted as which entities “win” and which entities “lose” due to a policy package. First-order cash flows aren’t where the money ultimately lands.

The proper simulation of policy effects on GDP (or jobs) is very difficult, because you must account for higher-order effects. Many models that attempt this do so poorly, because they have short time horizons (e.g. a few years or less) that don’t account for the re-integration of displaced workers into the economy and the long-term effects of efficiency gains on the economy. This is easiest to visualize in the context of an intervention (such as automation) that displaces workers and reduces spending while increasing productivity. In the short term, this sort of intervention looks bad, as it reduces jobs and GDP. But in the longer term, displaced workers find new places in the economy where they can work, and the entire economy produces more goods with fewer people. Many of the best policies to mitigate climate change, such as policies that improve products’ energy efficiency, improve material efficiency in industry, and promote renewable electricity generation, have disruptive short-term effects because they save money (reduce GDP). In the longer term, the economy will be better off for these reductions, just as the economy is better off today because of all the labor-saving and energy-saving devices and techniques invented since the Middle Ages.

Non-Financial Metrics

Fortunately, non-financial metrics are not as equivocal as financial metrics. For example, the Energy Policy Simulator includes an estimate of avoided premature mortality (human lives saved) due to reduced pollution. This is an unqualified good.

Similarly, avoiding greenhouse gas (GHG) emissions is an unqualified good, as it reduces both human suffering and huge financial costs in later years. There is widespread scientific agreement that the types of technological and policy interventions necessary to transition to a sustainable society are vastly cheaper than the damage that unchecked global warming would do to our economy.

Rules of Thumb for Policymakers

If the guidelines above regarding financial metrics make them seem too multifaceted to be used for policy guidance, a few rules of thumb might help: