A recent CPI study estimated the impacts of U.S. building energy codes on energy use and carbon emissions in new homes. While energy savings is a very logical place to start in measuring the effectiveness of these codes, it’s also important to understand their impact on costs. Much of the appeal of energy efficiency programs lies in the persistent finding that they are very cost-effective, often paying for themselves with their own energy savings over time.

We will be looking at cost impacts of the codes in more depth in the near future. However, we can do some simple calculations based on our energy savings results to serve as a starting point. In our study we found that, on average, a new household built in a state that has adopted a code at least as stringent as the 1992 Model Energy Code consumes 9.1% less final energy than a household built in a state without that has not adopted such a code. So, as a first approximation, we can assume that these households face 9.1% lower energy costs. The average U.S. household spent $2269 on energy in 2008, the most recent year for which measured data are available. 9.1% of $2269 is about $206; let’s take $200 as a nice round number. So, we estimate that the studied codes saved the typical code-built household $200 in 2008. We can eventually make this value more precise by considering different fuels separately and by paying attention to where in the country the code houses are being built, but it’s good enough to start with.

No one will argue with getting $200 in savings each year, but what are the costs of making buildings more efficient? Unfortunately, we are not aware of a definitive source for these costs. (If you are, do let us know! We’ll need to find or develop one eventually.) A couple of quick Google hits imply that the rough range may be something like $1000 to $3000 for an average household in today’s dollars, but we’re reluctant to put much weight on this. Also, understanding the cost increase to builders is only part of the story. If we’re evaluating things from the buyer’s perspective, we need to know how costs and savings are passed through to the buyer. If we assume that housing supply is perfectly competitive, then the capital costs are passed through 100%; however, this assumption may not be appropriate.

What we can do is think about how much of an increase in home purchase price the energy savings from the codes justify, thus sidestepping the question of who reaps the costs or benefits. To avoid speculating about future energy use trends and energy prices, let’s just assume that the studied codes generate a $200 annual savings in perpetuity. If we specify a discount rate, we can find the present value of this stream of savings: an amount of money today that we consider equally attractive to the perpetuity. The tricky part, of course, is the discount rate. If we look to consumer choices for energy efficient devices, we might choose a very high discount rate of, say, 30%. Discounted at 30%, an annual savings of $200 is worth only $667 today. However, there is plenty of reason to suppose that these choices are not simply expressions of the consumer’s time preferences, but are due in part to market failures such as imperfect information or to behavioral anomalies in decision-making. After all, a consumer that requires a simple payback on an investment may still choose to invest money in stocks and bonds instead of buying a more efficient refrigerator, even if the latter option provides a better return. So, perhaps we should use a discount rate equal to the rate of return available on the capital market. There is no one correct value here, and the return obviously depends on the riskiness of the investment, but 7% is a commonly chosen value and is the real rate of return of the S&P 500 since 1950. Discounted at 7%, our savings are worth $2857 today. Or, since decisions about energy efficiency in buildings have implications for society (they affect the need for energy infrastructure, which is funded by public dollars, as well as health and climate impacts from energy generation), we might choose a lower rate rate more in keeping with the literature on social discount rates. At a 4% discount rate, our savings are worth $5000. At the 1.4% discount rate used by the Stern Review, our savings would be worth $14,286.

Clearly, as is so often the case, the choice of discount rate is consequential. As a more concrete way to look at the issue, let’s consider my colleague Andrew’s suggestion. We have a pretty robust market for financing housing purchases: the mortgage market. (Insert requisite joke about robustness of U.S. mortgage markets here.) So a natural discount rate to use for this problem is the mortgage rate (plus property tax). In other words, we’re asking how much receiving $200 each year ($17 per month) will allow you to increase your loan. These days, mortgage rates hover around 4%. We’ll add on 1% in property taxes. We assume a 30 year fixed rate mortgage with a 20% down payment. Using these values, we can calculate that $200 annually in energy savings would allow us to qualify for a house that is worth an extra $3843. Of course, this requires an additional outlay of 20% of that number for a down payment; if we prefer a value net of this down payment, we get $3074. Granted, mortgage rates are currently at historic lows. If we rewind the clock ten years and use a 7% mortgage rate (plus the same 1% property tax rate), we get a $2814 increase in allowable home price, or $2252 net of the increased down payment.

Where does this leave us? Well, there’s no one way to answer the question of whether codes save money; it depends on the way we discount. And we don’t know exactly what the increased cost of building to code is, nor who bears the added construction costs and reaps the energy savings. Our preferred way of discounting (referencing the mortgage market) suggests that codes are probably more than paying their costs. As a simple way to think about it, the average 1992-2008 state energy code allows you to finance a home worth several thousand more dollars through the savings it provides; if the cost of building to code is less than this, the home buyer is winning.

More on this coming soon. If you have thoughts or questions, please leave a comment!