Skip to content

The Map Is Not The Territory

This post is by Phil Price, not Andrew.

My wife and I are building a new house, or, rather, paying trained professionals to build one for us. We are trying to make the house as environmentally benign as we reasonably can: ducted mini-split heating, heat pump water heater, solar panels, heat-recovery ventilator, sustainably harvested lumber, yada yada. But there’s one exception: we insisted on a wide expanse of north-facing windows that look out on the creek that runs through the backyard.

All those windows cause a problem because even low-emissivity windows are much less insulating than the insulated walls they replace. We have compensated by making the walls extra-thick, and the ceiling too — indeed we have to extend all of the windowsills and other window trim because the walls are thicker than the thickest pre-fab windows; ditto for door jambs etc. And yet, even with all that extra insulation, we were still not quite compliant with the California Title 24 energy requirements. This is entirely because there is a large penalty for using a heat pump water heater rather than a gas water heater, which makes no sense whatsoever given California’s energy mix and which everyone expects will be revised in the next set of Title 24 requirements, but you go to war with the regulations you have, not the regulations you want.

How does one determine whether one is in compliance with Title 24? One runs approved software that evaluates a computer model of the building: the orientation, the insulation values of the walls, how big are the windows and what kind are they and which direction do they face, and so on. The model said that our original design was just barely non-compliant, even with its heavily insulated walls and ceiling. We could take out some of those precious windows, or reduce them in size, but what other options did we have (besides hooking up to the gas pipeline and using a gas water heater)? The architect noted that we were compliant in all seasons except winter, so even a little bit more solar gain would help. Why not increase the size of the south-facing windows (which face the street), which would get us credit for enough winter heat gain that we would be in compliance? I pointed out that since the south side of the house is in the shade in winter, due to the large house on a hill immediately to the south, increasing the size of the south-facing windows would make the winter energy performance worse, not better. The architect agreed that yes, it would make the energy performance worse, but it would make the predicted energy performance better, and that’s what we need in order to get the permit. The suggestion bothered me some, although I know I shouldn’t find it surprising: most people think of regulations as hurdles to be cleared, and if it’s cheaper or more convenient to honor the letter of the law rather than the spirit, that’s what they’ll do.

I don’t really know what alternative to suggest for enforcing energy efficiency requirements (whose goals I support). Classical and neoclassical economists will say you can just charge more for energy and if you get the cost right people will automatically make the right energy choices, but those economists make their evaluations based on a model world that differs from the real world just as the model of our house differs from the real house. Anyway the whole experience just made me wish — not for the first time, or the last — that people could be counted on to do the right thing so we wouldn’t have to have all these damn rules, especially when they are actually counterproductive in some cases. But they can’t, so we do.

I was going to give two more examples of people confusing the model with reality that have come up in my work as an energy industry consultant, but I think this post is long enough and anyway it’s basically the same story with slightly different characters. The point is that optimizing the performance of the model of your building, or of your electric grid, or of your demand response program, will not optimize the actual performance of your building, or your electric grid, or your demand response program. Sometimes the difference doesn’t matter much but sometimes it matters a whole whole lot. Never forget it.

This post is by Phil Price, not Andrew.


  1. 3rdMoment says:

    This interesting ecoomics paper found that the “predicted” energy savings from a weatherization program were more than three times larger than the actual savings observed in the field.

    Perhaps consumers who are failing to make the “right” energy choices are not so irrational after all?

    • Agree, regulations should be designed to do better on average or overall and allowing too many exceptions may put this at risk.

      Like in statistical reasoning, individual cases seldom really matter, only what repeatedly happens.

      • Rahul says:

        Isn’t this a great example why the regulators should not try to go into the specifics of evaluating implementation?

        It’s always going to be difficult to come up with a regulatory code model that precisely accounts for all local factors and also be in sync with how technology changes eg heat pumps.

        Instead, the regulator should just factor in the penalty into gas, water and electric utility prices ( pigouian tax?) and no one has an incentive to construct energy sub optimal houses.

    • Andrew says:


      Thanks for the link. This story reminds me of the general issue that published effects tend to be overestimates (the “decline effect”).

      Ironically, one of the authors of the linked paper is Michael Greenstone, who published that notorious paper that used a biased procedure to estimate the effects of air pollution.

    • jim says:

      Ooo! 3rd, nice work. I love this bit at the end of the abstract:

      “Even when accounting for the broader societal benefits derived from emissions reductions, the costs still substantially outweigh the benefits; the average rate of return is approximately −7.8% annually.”

      This is another way of saying that the actual prices of the various technologies accurately reflect their long-term costs: in dollars, in resources – and even in the cost of purported “externalities” like emissions.

      IOW: price-coordinated economies work.

    • Phil says:

      If you’d asked me to predict the outcome of the study, I would have guessed the energy savings were roughly half what was predicted; that’s in line with what I’ve seen before for weatherization. I’m not shocked if it’s only 1/3 in some particular program or programs, but it’s even lower than I think is typical. It’s both funny and frustrating that disappointing weatherization results have been regularly reported for (at least) 30 years but people don’t seem to adjust their estimates downwards in response. The usual explanation — which may in fact be true — is that the estimates are based on what is seen when dwellings are weatherized by skilled people who know that they are part of a study and are therefore very careful, whereas the retrofits in the program are performed by people who are just trying to check a box so they can get paid. Ah, funny, I just searched for ‘disappointing results insulation’ and one of the hits is a book from 1985, “Energy Efficiency in Buildings: Behavioral Issues”, by Paul Stern. It says “Gaps and bypasses in insulation in attics can account for 35 percent of a home’s heat loss in winter (Beyea ,Dutt, and Woteki, 1978). They are a likely explanation for disappointing results in retrofits because small, hard-to-reach attic spaces may easily be passed over by an insulation contractor who is in a hurry or who is not careful.”

      I guess there are different ways to assign blame, if that’s the goal. It’s ridiculous for regulators or researchers to fail to learn from decades of real-world experience when assessing the likely effectiveness of a proposed program. This reminds me of and is also an example of mistaking the map for the territory: the model says you should get X% savings, but the model is not the world. But also, if contractors are in fact doing a less-than-thorough job, that seems like a moral problem there too.

      I live in a house that was built in 1918 with no insulation. I insulated the attic myself, but I know I did not seal all those problematic little gaps and bypasses. And we have only added wall insulation when we are doing some other work on the wall, such as reshingling or removing a chimney, so two of the four exterior walls are still uninsulated. I live in a very mild climate so this isn’t nearly as bad as it would be in most of the country. You can bet that if we lived in Minnesota we would have insulated! But we would not have expected to get the theoretically achievable effectiveness of the insulation.

  2. Daniel H. says:

    We had comparable discussions when building our house and I did discuss energy efficiency with many people since then. What has struck me is often there’s a large discrepancy between subjective truths (things we expect to be true) and calculatory truths (what you find when doing the math). For most people, including myself, going by the math is (averaged over all decisions) way better than doing subjective optimizations, even when some cases differ.
    (This topic might relate to the general theme of decision theory on this blog… somehow)

    • Phil says:

      I’m not sure how or whether this is related…actually I guess it isn’t but I thought you might appreciate it anyway: We are required to put in lots of insulation and to use low-E windows and so on (or at least we are effectively required to do so because there’s no other way to meet the Title 24 requirements). I don’t really have a problem with that. But I asked the architect: with all of this these windows and doors and air barrier etc., the house is going to be extremely airtight. How do we get enough ventilation to provide decent air quality? And she said “oh, that’s easy, we just set the bathroom fans to run all the time.” She said that is normal and she wouldn’t even have mentioned it if I hadn’t asked about it.

      This makes no sense whatsoever: go to all this effort to make the house airtight, and then run fans all the time to defeat the airtightness, WTF?! I said no, that’s ridiculous, we’ll put in a heat-recovery ventilator (HRV), so we did. It’s not a trivial expense but it will pretty quickly pay for itself compared to exhausting conditioned air (heated or cooled, in heating or cooling season) all the time. Amazingly-to-me, neither the architect nor the general contractor has any experience with HRVs, and even the HVAC subcontractor says they’ve only done about five in the past five years..which means most people really do build these super-tight houses and then either severely-under ventilate them or else deliberately exhaust conditioned air all the time!

      • Terry says:

        “most people really do build these super-tight houses and then either severely-under ventilate them or else deliberately exhaust conditioned air all the time!”

        Under-ventilation is pretty scary in new construction where many building materials outgas for years.

  3. Jay Livingston says:

    I hadn’t thought about Korzybski since high schoool, and that was a long long time ago. Thanks.

  4. qedrr says:

    === ” I don’t really know what alternative to suggest for enforcing energy efficiency requirements.. ” ====

    That’s a highly political/ideological statement, not some economic observation.

    “Energy Efficiency” and “Economic Efficiency” are differnct concepts.
    If a homeowner is struggling to make the monthly mortgage payment and other bills, big upfront investment in home energy improvements is rationally a low priority; most people can not afford the nifty energy technology discussed above.

    “requirements” for energy efficiency are a very arbitrary dictate from somebody… with some mechanism of forcing others to comply.
    Phil seems concerned about inadequate ‘enforcement’ mechanism … and not the larger issue of economic tradeoffs for the general populatiion.

    Why do politicians in Sacramento get to decide energy efficiency requirements for tens of millions of people in all manner of economic situations?
    This is a political issue.

    • Phil says:

      “Why do politicians in Sacramento get to decide energy efficiency requirements for tens of millions of people in all manner of economic situations?” Trivially, it’s because the state and federal constitutions set up rules by which elected legislators pass laws, and those legislators have chosen to make laws that include energy efficiency requirements. No surprise there. Perhaps you mean why _should_ politicians in Sacramento get to decide…that’s a question of political philosophy and moral philosophy that is well beyond the scope of this post!

      The point of this post is to emphasize the difference between the models we use and the actual systems we are modeling. Those differences can be quite important.

  5. Frank says:

    You write:

    >Classical and neoclassical economists will say you can just charge more for >energy and if you get the cost right people will automatically make the right >energy choices, but those economists make their evaluations based on a model >world that differs from the real world just as the model of our house differs >from the real house.

    It would be more accurate to say “the overwhelming consensus of economists regardless of political leanings says” see for example this petition:

    A carbon tax won’t result in everyone automatically making the right decisions. What it will do is create a positive feedback loop that links the choices of people and businesses to their economic consequences. If people aren’t sufficiently responsive to the tax, just keep increasing it. There’s a price at which people and businesses will find it worth their while to do precisely the calculations you describe above. The difference is that their incentives under a carbon tax will be to reduce emissions rather than meet an artificial, model-based target. Even better: if you really value that north-facing view, you can cut back on carbon elsewhere in your life to compensate for it under a carbon tax, but not under command-and-control regulation.

    • Frank says:

      Typo: I meant to write

      “What it will do is create a positive feedback loop that links the choices of people and businesses to their *environmental* consequences.”

    • Brent Hutto says:

      A dollar in carbon tax to me is not the same as a dollar in carbon tax to a 1%-er nor is it the same as a dollar in carbon tax to an impoverished family.

      Nor do myself, the 1%-er and the impoverished family have the same degree of choice available in making economic decisions.

      A large enough carbon tax to dissuade me from doing pretty much darned near anything I like without regard to carbon footprint might easily consume a huge portion of an impoverished family’s income.

      Imagine a single mother earning $20,000/year by working three jobs that she is able to hold because she can travel between them in a 25 year old car in which she drives 250 miles a week. Her 12,000 miles a year in a (likely) polluting, fuel inefficient cheap vehicle ought to reasonably be carbon-taxed far higher than the 15,000 miles a year I drive in my Prius. Where is she going to get the money if her yearly budget now includes a significant (to me) degree of carbon taxation…maybe something like $1,200 a year?

      Our economic system would make short work of deflecting the real cost (not in dollars and cents but in standard of living) of a meaningful carbon tax onto those at the bottom of the economy with limited flexibility to make carbon-saving choices. That sort of thing is one “feedback loop” that we excel at, as a society.

      • It’s useful to realize that there is some level of consumption required to maintain biological requirements. People need food, and to be in temperatures that are neither too cold nor too hot, and to keep children from wandering into traffic, and etc. This is true regardless of whether those people have employment opportunities in their area, whether they are physically or mentally disabled, whether they are male or female, etc.

        The consumption of these resources doesn’t go away unless the people die. It particularly doesn’t go away when someone else pays for them through complicated government housing or food programs. In fact, those programs consume more resources than if you just taxed people like you already are to pay for the housing or food programs, but instead of running the programs you just gave people enough money to buy the resources themselves. And that doesn’t distort the information carrying function of the price structure…

        So, the solution to the problem of dealing with externalities by taxing the consumption of resources that produce them is simple: use the taxes to fund universal basic incomes so that there are no people who are no “single mother earning $20,000/year by working three jobs that she is able to hold because she can travel between them in a 25 year old car in which she drives 250 miles a week”

        It turns out we could do this relatively easily in the US, we have enough resources to ensure that everyone has access to a minimally safe living environment and sufficient food, and can use their after-tax wage income primarily to buy either upgrades in quality of living environment and food, or additional resources they might want to consume.

        Doing this could dramatically improve the efficiency of our economy, and also dramatically improve living conditions for tens of millions of impoverished people… if only we could overcome the confusion people have between real dollars, nominal dollars, redistributive cash flow vs consumption, ideas about “progressive tax codes” being necessary for fairness etc…

        I’m not holding my breath for a dramatic improvement in economic and numeric literacy.

        • One particular foolish argument that is nevertheless prevalent is that someone takes the population of the US, say 320e6 people and guesses at the dollar amount needed to buy some basic level of security, lets say $1000/mo per person, multiplies by 12 mo/yr, and comes up with the idea that “this will cost 320e6*1000*12 = 3.84e12 = 4 Trillion dollars per year”

          Now I take a fat stack of $100 bills, let’s say 50 of them, and I hand it to you, then you hand it to me, then I hand it to you. We can do this say once per second. Evidently this activity produces $5000 a second. If you and I each do this all day long for a year we will each have an income of $1.58 Trillion, so just the two of us alone can pay for the whole thing! ;-)

          That’s sarcasm in case it’s not obvious. The point being that when you are redistributing 4 Trillion dollars a year it doesn’t mean you are consuming 4 Trillion dollars worth of additional resources. This is a concept average people just can’t seem to wrap their head around without a lot of hand-holding.

          Let’s assume that the two money-passers are upper middle class householders. Evidently their income is about $100k/yr each, and they give up whatever they were doing to pass the phat stakz, so the cost of their $3 Trillion/yr phat stakz game is actually $200k/yr because that’s how much poorer the world is because they gave up doing their jobs to pass the cash… (all costs are opportunity costs)

          The same concept applies to the UBI idea. Sure, you have to MOVE A LOT OF MONEY around, but moving money is just electrons in bank computers. It doesn’t consume much resources at all. Actually you undoubtedly consume less resources under UBI to produce the current situation than you would currently. The actual cost of UBI is very likely NEGATIVE relative to our current system. In the UBI system we wouldn’t have to give up whatever the bureaucrats who run the system could be producing, we wouldn’t have to give up the hundreds of hours a year each recipient of government programs has to give up in order to qualify and remain qualified. People wouldn’t have to give up working to keep their housing or child care benefits, poor people would have better financial situations and be less likely to purchase goods inefficiently (like single toilet paper rolls at the gas station vs a large family pack at the local Target), we’d have fewer emergency room visits, and fewer bankruptcies, and reduced petty crime, and etc etc.

          And, prices could better reflect consumption, including taxes on goods with externalities such as the polluting vehicle originally mentioned without imposing life-threatening burdens on poor people.

          • Dale Lehman says:

            Yes, people confusing efficiency and distribution all the time. Carbon taxes are (arguably perhaps) more efficient than most of the alternatives. However, the resulting prices become quite transparent, thus it is easy to object that the poor person (with 3 kids who drives 100 miles per day, or whatever) might not be able to afford the gas with the tax. It seems as if command and control policies, such as pollution control requirements on vehicles, is more responsive to the needs of the less-well-off. But this is mostly illusory – the costs of the equipment will affect the cost of vehicles and may not be any more affordable to the less well off. In fact, it often will be more expensive. And, the loss in efficiency (by choosing the wrong technology, failing to recognize substitution possibilities, etc.) reduces the size of the economic pie available to all. If that loss is small, it may not matter much, but the larger the inefficiencies, the more we should all care about it.

            Economists have a tendency (a fatal flaw in many cases) to ignore issues of distribution – they are messy and seem unscientific. Is a carbon tax more unfair than CAFE standards on vehicles? Analyzing a question like that is not trivial, but might be worth a serious effort.

            Daniel has always like the UBI idea – I sort of like it myself. There are many simpler ways to enact public policies than the ones we currently use. How about replacing our crazy private health insurance system with a more straightforward single payer system? Or replacing our ridiculous tax code with a simple flat percentage income tax beyond a minimum income level (with no deductions, credits, etc.)? I tend to like these approaches – I think they end up being fairer as well as more efficient. But there is a reason why they are not the way we do things. Politically it is hard to garner the support. And, the politicians don’t see much benefit from these alternatives – financing election campaigns requires convoluted tax and incentive schemes.

            So, back to Phil’s original post. To promote energy efficiency, carbon taxes seem an attractive option to me. It is not that I assume people will just “automatically make the right energy choices” but I think they will get closer than our indirect approaches to make the choices for them (by regulating the kind of windows they install, the type of vehicles they can drive, etc.). The latter is fraught with opportunities for mistakes, deception, political lobbying and other efficiency-sucking activities that serve private interests at the expense of public ones. The tax approach is not perfect, but let’s not let that be the enemy of the good. And, yes, let’s earmark the tax receipts into serious redistribution so that the most unfair aspects of using prices don’t further impoverish the poor.

            • BTW Dale, I consider a flat tax to go hand in hand with the UBI. Economists like the idea of incentivizing investment rather than consumption, so I see about a 25% flat federal income tax and a 10% flat federal sales tax as a good way to do that… those who save and invest don’t pay the consumption portion. Combine it with the UBI and the efficiency gains and improved price transparency are SUBSTANTIAL.

              • GDP per capita in the US is around $60k/yr, flat tax of 25% on income would produce about 15k/yr/person, another 10% consumption tax might produce another 6k, put together this is 21k/yr/capita.

                Now, we’re paying everyone in the country say 12k/yr under UBI, so 21-12 = $9k/yr left over after redistribution. Multiply by 320M people and the federal government would have about $2.9 Trillion/yr to spend on consuming things.

                Ignoring redistribution they’re already doing, the federal budget consumes $0.66 Trillion on military, 0.39 T on debt interest, and about a Trillion or so on stuff like the VA, agriculture, education, security, NASA, NIH, Census, etc etc.

                Which means that the $2.9 T/yr would cover the actual govt consumption.

                Sure, there would be winners and losers here, social security and medicare recipients might receive less, but children and poor families would receive more, and the efficiencies would be huge, so the pie would get bigger more rapidly.

    • Phil says:

      I’m a huge fan of taxes and other mechanisms so that prices reflect the societal and environmental cost of the goods and services we use. I think that is often, maybe even usually, far far better than trying to achieve similar ends by setting fuel economy standards and energy efficiency standards and so on. Bring on the carbon tax! I did not mean to imply that market mechanisms aren’t useful and important. I’m a big fan! I just think too many economists think their models are closer to reality than they are.

    • jim says:

      The biggest problem with any model involving carbon taxes is the mistaken assumption that they will be based on carbon. In reality, they won’t. Politicians will raid the cash stream at every possible excuse for every possible thing under the sun, dramatically distorting the incentive.

      Whatever the case, if wind and solar are becoming economic, then happy day, we don’t need a carbon tax.

  6. AllanC says:


    You might like/dislike this bit from Joe Lstiburek about how computer simulations are gamed to get buildings to pass energy standards (see 15:05 to about 16:00):

    The entire lecture is pretty good if you’re actually in to building science / want to know how to build a proper building enclosure.

    • Phil says:

      I’m very familiar with this general issue: I used to work at a government research lab and one of my research projects involved comparing a building’s actual energy performance to its predicted energy performance. I became very aware of…well, of what is discussed in this talk (thank you for passing it along. I can’t decide whether I like this guy’s manner or dislike it…either way it is entertaining and informative). I’m certainly aware of all the gaming that goes on to get LEED certification and so on. It is indeed disappointing that so many people don’t care whether or not they have an energy-efficient building, they just want the certification, and that there’s a whole industry devoted to making that happen.

      • Phli says:

        Actually, not only did I work at a government research lab, but it was the same lab (Lawrence Berkeley National Laboratory) where Lstiburek gave that talk. But I didn’t go to his talk. At that point I was only just starting to work on building energy stuff, and I do data analytics that is pretty far divorced from what he was talking about. His stuff is all a lot more interesting to me now that we are building a house.

  7. jim says:

    If you want to be green, do what’s most cost effective because after you look at the entire chain, that’s going to have the lowest resource impact. That’s what prices do: they communicate the most efficient consumption of resources which is ultimately going to determine the environmental impact.

    • Phil says:

      I’m a huge fan of price signals. My knock at economists is because I think that as a discipline they seem to be way too quick to believe their models reflect reality — they confuse the map with the territory. It’s not that I think they are wrong about prices being important.

      • jim says:

        I think people – economists, people in general – don’t grasp the complexity of the problem. So they make simplifying assumptions, well that’s normal in science, but if you make too many you can choose your favorite flavor of assumption and get the most pleasing result – and ultimately forget that you made any simplifications at all.

        Price is beautiful because intrinsically it cuts through all of that.

  8. Terry says:


    Is your solar installation subsidized in any way?

    Would it be cost-effective if it weren’t?

  9. Terry says:


    Sometimes I hear people say something like “the average American suburban home uses as much energy as 30 African Villages”.

    I don’t know if that is true or not, but let’s say it is for now. Do you have any idea how much you reduced your energy usage in units of African Villages? Do you only use as much energy as 15 African villages? 10? 25?

    If you got rid of all your creek-facing windows, how many African Villages of energy would it save?

    • you’ve got the units wrong, the question is power consumption: energy per time. But let’s say it’s equal times, the average american house uses as much energy in a year as N African villages use in a year… let’s do a Fermi problem to figure out N to see if you’re anything close to right.

      First off, Africa has a population around 1.3 Billion, but we shouldn’t include northern africa because that’s probably not what is being thought of in the supposed comparison. Let’s just go with western africa, which has about 400M people, and the US has about 320M, which to a physicist is the same number.

      Now based on the graphs here west africa uses around 0.5 tons of CO2 per capita per year

      They don’t mostly use it for electrical generation, but that’s irrelevant, energy is energy. You’ll notice that the electricity generation ranges over orders of magnitude, but the greenhouse gas emissions are all similar order of magnitude reflecting the fact that if you’re not using electricity you’re using fuel or some other thing.

      Now, lists GHG emissions per capita, with US having about 16.5 tons/capita.

      A typical house in the US has say 3 inhabitants so we’re talking about 16.5*3 = 50 tons/house/yr

      This is equivalent to about 100 west African people. If you figure a “village” is maybe 100 to 1000 people, one US house is somewhere between 0.1 and 1 village of consumption rate.

      So, you’re off by a factor of between 30 and 300. But “village size” is a pretty arbitrary thing, so the real point is a family of 3 in West africa uses around 1.5 tons of GHG per year, and an american family uses about 50. Which is a ratio of about 33, which is close to the numerical value you gave. So, no, houses don’t use “30 times what an african village uses” but may use about 30 times what a rural African household uses.

      • Terry says:

        The 30 number wasn’t mine, it was from some activist or other. Your analysis sounds about right for the average African. The fall back interpretation is that “there exists villages such that they use 1/30th of the average suburban home. Note also that “average suburban home” is probably above average in energy consumption.

    • Phil says:

      Do you mean how much did we reduce the energy usage by adding that marginal amount of insulation that took us from non-compliant to compliant? Or do you mean how much did the totality of everything we did reduce it from some sort of minimally acceptable house, like maybe the very leaky and un-insulated house that this new house is replacing? I don’t know the answer to either question, but they are clearly very different!

      Since I really have no idea of the answer to your question, I’ll answer a totally different question. I think the house will generate between 60% and 80% of the energy it uses…if we don’t count the energy for the electric vehicle that will be charged from the house. We will probably use something like 9 kWh per day (or, if you prefer, an average of 375 W) just to charge the car. Certainly we will be waaaay below a typical U.S. suburban house: we live in a very mild climate (Berkeley CA) and the house is only 1550 square feet of conditioned space, so even if it weren’t for the other energy efficiency stuff it would not use a lot of energy by US standards.

  10. Terry says:

    Related Fun Fact:

    In California there are benefits to “renovating” a house rather than demolishing the house and building a new house. I think it is a tax thing.

    Over the years, the definition of “renovation” has been loosened to the point where you only need to keep one wall from the original house. So a lot of homeowners have demolished all but one garage wall of a house, encased that wall in a brick or drywall covering, then built a completely new house around the wall. Voila, it’s a renovation.

    Moral: well intentioned regulations are often relentlessly gamed to the point of absurdity.

    • Phil says:

      I live in California. I’m not aware of any such “one wall” rule — I don’t think there is such a rule — but there are certainly rules that are approximately that ridiculous. The ones I’m aware of are based on a “50%” rule, not a “one wall” rule. For instance, if you are doing a renovation that costs more than 50% the value of the house, you are required to bring the house up to current building code in certain ways. But at 49%, hey, no problem, leave it the way it is. There are several things like that.

      I think it’s in The Affluent Society (by Galbraith) that the author has a little discussion of how laws tend to get overcomplicated, using an example of laws about providing bathrooms for workers. People have to go to the bathroom sometimes, so if you hire someone you have to give them a safe, sanitary place to do so…that makes sense, right? So you start with a law that says workers must have a safe, sanitary place to go to the bathroom. But someone provides such a place a long way from where the workers are actually working, and makes the workers go off the clock when they go to use it. So you amend the law to say the bathroom has to be reachable within a certain amount of time. But then these corner cases crop up. What about fighter pilots? What about cowboys on the range? What about this, what about that…before you know it you have thirty pages of regulations about bathrooms.

      Hmm, this reminds me of Temple Grandin’s checklist for slaughterhouses. (If you don’t know who she is, she’s worth looking up). As with the paragraph above, this relates to a book that I read many years ago so I may get some details wrong. My recollection is that there are lots of regulations around slaughterhouses to try to make them not-too-inhumane for cows, and fairly safe for workers. For instance, they don’t want cows falling on slippery floors, so there are regulations about how much slippery the floors can be. Stuff like that. As I recall, Grandin pays no attention to whether a slaughterhouse does or doesn’t comply with the regulations, she has a much smaller set of items, maybe just five or ten. Things like “Does fewer than one cow in 500 fall down?”, and “Does fewer than one cow in 100 bellow in distress” (I made up those numbers, but you get the idea).

      • Dale Lehman says:

        The bathroom story reminds me of this: I testified in a regulatory case in Texas. One of over 1000 issues that the Public Utilities Commission needed to decide (in a 3 day hearing) was what rate should company X be able to charge company Y for its employees to go to the bathroom in a facility where they collocated (allowed by law) some of their facilities? Company Y claimed there should be no charge (it is their employees “right” to go to the bathroom) while company X assessed a charge on the basis that many such facilities were not staffed and so an escort would need to be provided in order to protect their own facilities against damages.

        I don’t recall the outcome, but I’d wager it was about halfway between the two positions. Life as a regulatory economist was fun, unless you actually thought about what you were doing.

      • Terry says:

        To be honest, I was being lazy and someone I know told me very earnestly about the “one wall” rule. Googling, I couldn’t find support for it. I should have known. My informant has not been very reliable on other matters too.

        • Phil says:

          Terry, It’s my experience that when people complain about ridiculous regulations the regulations are often not as ridiculous as they make them out to be. Don’t get me wrong, they may very well be ridiculous, just not as ridiculous as people make them sound. Your ‘one wall’ example is more ridiculous than the actual regulations, but the actual regulations are still kind of ridiculous. In the vernacular of ten years ago, we would say your claim isn’t true, but it’s truthy.

        • I’ve physically seen these single walls left standing during total rebuilds, so I think this is a real hack that is used to get around certain regulations… but I suspect it’s probably at a county level or something rather than statewide.

          Definitely seen them in LA county, might have seen them in Contra Costa County, not sure.

          • Phil says:

            It’s possible. Here in Berkeley it’s sometimes the case that half of the exterior walls are left standing (by area); I’ve heard that that’s because of a 50% rule of some sort, though I haven’t actually seen a rule like that. But I know I have seen just a couple of exterior walls left. Not just one, that I can recall, except where a building facade was being preserved for aesthetic or historical reasons (which is the case for some construction right now in Downtown Berkeley).

          • Terry says:

            The plot thickens.

          • Terry says:

            Aha! I may be lazy, but I’m not as wrong as I thought I was.

            This explains it in some detail. Weirder and more complicated than I thought.


            • Phil says:

              Terry, that is pretty much as you described it! That is indeed ridiculous. I mean, it’s one thing if it’s a real rarity and it’s just not worth changing the laws to deal with 0.05% of renovations per year or something. But if this is a regular thing they should change the regulations. The current rule does nothing but increase the cost, it does not help preserve historic buildings or the ‘fabric of the neighborhood’ or anything. Eh, well, actually that may not be true: if you could do a genuine renovation for $800K, or tear down the whole house and build a new one for $1M, maybe you build the new one. But increase the cost of the new one to $1.3M and maybe you choose to renovate. So maybe this does help preserve whatever people are trying to preserve. But in that case they could do the same thing by simply charging $300K if you want to tear down the house, and avoid the theater with the walls.

            • Yeah, it’s not just historic stuff though. I’ve seen it with everyday renovations, even one a couple blocks from me and there’s no historic district here or anything. This appears to be just straight up regulatory capture.

              • Anoneuoid says:

                This place perplexed me for awhile, then I found out it was due to some perverse incentive due to a poorly thought out law:

              • Terry says:


                I feel culturally enriched just looking at it. Totally worth it.

              • AllanC says:

                Phil, Daniel, Anon,

                I live / build in Ontario and am unfamiliar with the California building process. However, in many municipalities in Ontario you typically do not have to comply with the latest lot setbacks if you maintain the existing foundation. This is sometimes useful because today’s setbacks are typically much more stringent then they used to be. For example, cottages used to be built very close to or even on the water. Now most municipalities require a 100LF setback from the waterfront.

                Anoneuoids pictures shows one wall very close to the municipally owned sidewalk. It’s possible there is some similar regulation in California that permits use of existing setbacks if some sort of building element is maintained on that setback (in this case a wall).

                Then again, maybe not. Building departments can be funny.

              • AllanC says:

                My mistake I viewed the picture way too quickly on my phone. It is not near a municipal sidewalk but the same idea would apply if the property boundary is near where the picture was taken.

          • Martha (Smith) says:

            There was a “one wall” rebuild-called-renovation in my neighborhood a few years ago. There seemed to be some hanky-panky along the lines of the owners brought in Habitat for Humanity to take usable house parts for them to use either in their buildings or in their resale shop, but didn’t tell them the rules, so the result was one wall remaining, and much larger house than was described in plans being built.

  11. Jeff Walker says:

    Phil — if you’re not familiar with you should look at the site.

  12. expr says:

    For the north facing windows, would some sort of insulating shade or cover that increased the window R-value change the rating?
    I can imagine something that folds off to the side when not in use
    I have used a chunk of coroplast (sp?) (plastic corrugated cardboard) in windows in apartments and it makes a big difference
    In one place with really bad windows I would get ice on the inside of the window (behind the coroplast) while it was high 60’s in the room.

    • Phil says:

      expr, I’d have to check the model but I’m pretty sure anything that requires the occupant to do something — like fold an insulating the shutter over the window — would not be taken into account in the model. Maybe if it was automated it would be allowed to count, I’m not sure. Basically the regulations assume a lot of people won’t do something just because it saves energy. I think that’s probably a decent assumption.

      Similarly, outdoor lighting (deck, exterior stairs, etc.) can’t just have a regular switch that relies on the occupant to turn it off. You have to have one of four options: (1) photocell and motion sensor, (2) photocell and timer, (3) astronomical time clock (meaning it knows when the sun comes up and goes down, it’s not just based on the hour), or (4) eh, 4 is basically like 3 only can be more complicated. Whichever one you choose, the point is that the light can’t be on for more than a few hours unless there’s someone there to make use of it, and they don’t want the light on during the day unless someone explicitly triggers it. No turning the light on at dusk and forgetting to turn it off, no leaving it on all the time, etc.

  13. Dzhaughn says:

    My evil twin says:

    Why does Phil have more faith in his model for valuing alternatives than he does in the economists’ model or the Title-24-architect-apparatchik model? They’re all false, like all models.

    The purpose of the ilk Title 24 isn’t just energy concerns–it’s also establishing precedents and norms for future architects, apparatchiks, and regulations. Hence things like the electric heat pump penalty–irrational, yes, but we got some serious concessions in exchange for dropping that! Such concerns matter much more than the marginal energy use of a custom house on the stream, but aren’t in your model. So just relax, get certified, and Support the Party.

    • Phil says:

      A point I often make is that people have to make decisions somehow. As Bill James (or Andrew) might put it, the alternative to a particular model isn’t no model, it’s a different model. So, sure, it’s always fair to ask why I think the model I’m using, which can be explicit or implicit, is better than the alternative.

      In this case that’s a really easy question, though. I assume I don’t have to explain why replacing some heavily insulated wall with some windows that are shaded in winter is not going to improve winter energy performance, but will instead make it worse.

      As Andrew and I have discussed (in print a few times, for instance in our decision analysis paper about radon remediation, and in private many times) regulations are often based on a very simplified view of the world, and they should be. Rather than an either/or decision about whether you get a speeding ticket, we could imagine that the size of the fine depends on the weather and the number of people in your car and the weight of the car and the exact speed and the amount of traffic and the type of road etc etc etc, since all of those effect the expected number of people you will injure or kill, but it’s not like it would be reasonable to have a law like that. One could say the model on which the speeding ticket is based is very simplified: below a certain speed you’re judged to be driving safely enough, and above it you aren’t.

      The energy model used by Title 24 is a lot better than either/or. If we’re going to have energy conservation regulations, it’s better to base them on a model like this rather than (for example) simply requiring a specific amount of insulation and a specific type of window and so on, because you could use all of the prescribed elements and still end up with an inefficient house, or you could skip some of them and still have an efficient house. Better to use a whole-house model, even if it’s oversimplified. And architects use CAD software anyway, so it’s not a huge burden to use the software. I don’t really know what alternative to suggest for enforcing energy efficiency requirements. That’s more or less what I meant when I wrote “I don’t really know what alternative to suggest for enforcing energy efficiency requirements” in the original post!

Leave a Reply