Jeff Biddle on migration and air conditioning

Here is the abstract of his piece “Air Conditioning, Migration, and Climate-Related Rent Differentials“:

This paper explores whether the spread of air conditioning in the United States from 1960 to 1990 affected quality of life in warmer areas enough to influence decisions about where to live, or to change North-South wage and rent differentials. Using measures designed to identify climates in which air conditioning would have made the biggest difference, I found little evidence that the flow of elderly migrants to MSAs with such climates increased over the period. Following Roback (1982), I analyzed data on MSA wages, rents, and climates from 1960 to 1990, and find that the implicit price of these hot summer climates did not change significantly from 1960 to 1980, then became significantly negative in 1990. This contrary to what one would expect if air conditioning made hot summers more bearable. I presented evidence that hot summers are an inferior good, which would explain part of the negative movement in the implicit price of a hot summer, and evidence consistent with the hypothesis that the marginal person migrating from colder to hotter MSAs dislikes summer heat more than does the average resident of a hot MSA, which would also exert downward pressure on the implicit price of a hot summer.

The pointer is from Ross Emmett in the MR comments section, very useful comments overall.  Biddle has two other pieces on the history of air conditioning, and Biddle has other interesting pieces as well, he is apparently an underappreciated economist.

Here Scott Sumner details the import of state income taxes.  In my view not the “main” factor, but a significant factor nonetheless, excerpt: “On the west coast, all states grew faster than the national average. Yes, its climate is nicer that the south central region.  But look at the more detailed data and you’ll see that hot and sunny Washington state and Alaska grew the fastest of five bordering the Pacific.  And oh by the way, Washington and Alaska are the only two with no state income tax.”  I’ll add this point: to the extent income inequality is rising, a relatively small number of cross-state migrants can lead to a noticeable difference in cross-state growth and job creation rates.  And the high earners are precisely those who are most able and most likely to leave a high-tax state for a low-tax state.


Air conditioning's role in the economic rise of the South is like the cotton gin's role: historically obvious, but not terribly relevant to recent trends.

The interesting current paradox for left v. right to furrow their brows over is that left wing state-level policies, such as in Vermont, are better at preserving Vermont for Americans (especially white Americans), while right-wing state level policies, such as in Texas, lure in lots of foreigners.

Today, the difference between left (e.g., Vermont and San Francisco) and right (e.g., Texas and Atlanta metro area) in America is mostly that leftist policies are expensive while conservative policies are cheap. Expensive leftist policies require need more productive white / Asian elites, while cheap rightist policies can accommodate the less productive Hispanic / black masses.

that makes no sense, didnt Reagan triple national debt and Bush Jr doubled it, and Clinton was the first president in decades to get a budget surplus? evidence points the other way around, conservative policies cost more. Why are you ignoring the evidence?

It's state and local level policies that matter. Blue states are very expensive to live in, they have regulated industries that charge more (such as not allowing competition in health insurance or car insurance), they have higher taxes and more restrictive zoning laws. New York State runs ads on TV trying to bribe businesses to come to NY, offering them no taxes for 10 years. Massachusetts uses similar bribes for specific companies and industries. Low cost states don't need to bribe companies.

Here's the ad:

I grew up in Los Angeles's San Fernando Valley, so I was very aware from the early 1970s onward that liberal and rich Beverly Hills and Malibu were much more enthusiastic about expensive environmental regulations than were conservative and middle class Chatsworth and Granada Hills. The basic issue is that life is pretty great already in the People's Republic of Santa Monica and the like, so why disrupt it with development?

Especially with Prop 13 freezing property taxes on property that remains locked in the past and in hereditary ownership.

Why would you split a beach front lot assessed as remote barren land with a cottage in 1980 for property taxes of $2000 per year when the split and redevelopment would result in taxes on two ten million dollar properties based on the fair market price of the two new properties?

TV ads and special tax deals for businesses that move are actually pretty marginal. If they land a corporate headquarters or two that's icing on the cake but ultimately no state in the US is able, IMO, to base its economy on 'stealing' business from other states by offering lower taxes and regulations.

Blue states are expensive to live in because you are living close to large, highly concentrated and diverse industries that offer lots of opportunities to get rich. Economic growth drives growth in taxes, regulations, zoning restrictions etc.

Simple question for those who are enamoured with the story of sunny, less regulated red states overtaking blue ones... If the driving theory here is that economic growth moves from high tax/regulations to low ones then how exactly did blue states get so many businesses and jobs to move in the first place? When exactly was NY, for example, the low tax, low regulation paradise and Texas was the opposite?

"low cost states don't need to bribe companies" REALLY???

Can you link to some data on this ???

"Low cost states don’t need to bribe companies."

Just completely, utterly wrong:

"The Times analysis shows that Texas awards more incentives, over $19 billion a year, than any other state."

I've always questioned such figures. $19B assumes that had Texas not provided the incentives, the businesses would have set up shop there and paid the full tax of $19B. But that can't quite be true. I'm sure at least a few businesses made the decision based on the incentive, so the actual cost of the incentive would have to be less than $19B. It could also be a net gain if Texas was especially savy, only offering incentives where it actually made the deciding factor. But that's probably asking for a lot. More likely a good chunck of the incentives either went to companies that were going to be in Texas anyway or were rewards for campaign support rather than economic growth, so I suspect it was a net cost to Texas...just less than $19B.

However consider the nature of the business. Some businesses have very deep roots and others do not. A break on property taxes for a few years is unlikely to motivate a deeply rooted company from leaving but would attract a company whose existence is more transitory and brief.

"it was a net cost to Texas…just less than $19B."

Well sure, but there is a vast distance between "red states don't have to bribe!" and "they bribe somewhat less than $800/cap."

@jon, it is a fair criticism that conservatives haven't been very good at reducing the size of government in practice. However, your one-sided analysis is just silly. The president doesn't have sole responsibility for the budget. Opposing Congresses were big factors in two out of the three examples you mention. And the increase in the debt under Obama and a Democratic Congress dwarfs any previous increase, except for major wars.

I agree with you Dan, except the President only proposes a budget.
Congress taxes and Congress spends.

Oh, and Jon, the person you want to thank for the surplus was Newt Gingrich.

The president usually has a large influence in this area, even though his formal role is small. I think it is fair to see it as an area of joint responsibility.

If you look at the federal deficit as a share of GDP, since WW II every Republican administration left office with a larger budget deficit than they inherited while every Democratic administration left office with a smaller deficit than they inherited.

I use administration so I can count Nixon & Ford as one..

You must be counting Kennedy-Johnson as one administration, too. Otherwise Kennedy would not follow the trend. And you must be using inflation-adjusted dollars rather than nominal, saving Jimmy Carter's bacon. Even ignoring these arbitrary decisions, you have a sample size of nine and are making inferences about differences between subgroups. I'm going to guess the p-value is not significant.

"size of government" is a pretty unclear concept. Are we talking about the amount of money gov't spends (in which case the deficit is irrelevant)? Are we talking about total taxes (in which case an increasing deficit would indicate Americans are getting more without paying more in taxes)? Are we talking about actual gov't spending on things like roads, police, military, NASA etc while excluding entitlement spending on things like Social Security, Medicare etc (not quite that silly considering that individuals decide how to spend their social security check while the gov't directly decides how operational budgets are spent)? Or are we talking about the burden of laws and regulations on individual and collective freedom (in that case the budget may be totally irrelevant to the discussion).

Libertarians would like to believe that this is all the same thing but it needn't be. One could have a very light gov't in terms of taxing and spending but filled with lots of trade restrictions. For example, in the 1600's you had almost no taxes and gov't programs but kings happily doled out special monopolies and licenses so one's ability to actually do serious business without gov't interference was quite limited.

Maybe jon will some day figure out that there is this institution called 'the United States Congress' and that the Democratic Party controlled the lower house thereof from 1933 to 1947, 1949 to 1953, and 1955 to 1995.

If 'leftists policies' are more expensive then why are Red states almost always net consumers of Federal tax dollars rather than suppliers?

right (e.g., Texas and Atlanta metro area)

If you take away the Atlanta metro area you also take away almost all of Georgia's economic success. But as Red as the state may be, it is the metro area that is almost certainly blue?

Likewise Texas's metro areas are rather blue compared to the state as a whole and, I suspect, the main drivers of the state's economic power. The only exception with Texas might be energy but that's simply the luck of where the oil fields ended up...unless you want to argue the dinosaurs opted to die on land they knew would one day be solid Bush country!

Being a net consumer or supplier of federal tax dollars has nothing to do with cost of living in an area. If anything, I'd imagine it's negatively correlated (people in expensive areas pay more tax, consume less direct government services).

To be fair high cost NY and CA are slightly net takers of Federak tax dollars and TX is a slight giver, but not by much. The rest of the South and most of the remaining Red States are net takers despite being supposedly low cost places to live. TX, though, might be an anomally because of oil royalities.

I think the error you are making is that by definition a high cost of living area means a high income area and it is almost always more attractive to live closer to a high income area than a low income one in terms of opportunity. By that consider a young writer. It is more attractive for him to take a tiny flat in the bad sections of NY or LA in order to be close to the high income areas than it would be for him to enjoy a much larger space in Kansas. Hence high income areas are going to always be higher cost of living areas because demand will bit up prices. NYC was a higher cost of living area, I'm sure, back in 1910 long before any onslaught of evil leftist programs and taxes.

I think that's exactly Sailer's point -- being wealthy leads to enacting various "blue" policies. Like an HOA adopting a bunch of petty rules to try to keep the riff raff out.

Right. Consider liberal Malibu, with its 26 miles of coastline for its total population of 12,645. Rob Reiner and Barbra Streisand aren't going allow any development that would make it easier for anybody who isn't stratospherically rich to move to Malibu. Heck, Reiner just organized a successful referendum banning the building of a Whole Foods on the huge vacant lot (maybe 100 acres of flat, undeveloped land) in front of the Malibu City Hall.

Given the Federal deficit, _most_ states are net consumers of federal tax dollars, blue or red.

This post and the prior post got me thinking about the role of indoor heating in attracting people to the North (specifically the Northeast). Even when it's below zero, I can still be cozy by my propane fireplace. Of course, if I do have to go outside, I have to put on a heavy coat, hat, boots and gloves. If I am walking my dogs, I may also need to strap on my cleats for a walk on icy roads. However, spring, fall and summer are very nice here in NE Pa, so the winter is the price we pay for the other three seasons.

I have never spent much time in the south during the hot summer months, but the tradeoffs are similar - beastly summer versus nice spring and fall and cool winter. And those who live in FL, for example, find the winter cold; when we have traveled to Florida and go in the pool in February, the natives know right away that we are from out of town. Air conditioning certainly makes the summer more bearable but being outside is something to avoid - like winter in the NE. You can travel up and down the East Coast and get a different balance of seasons - live in NC and you get a bit of wintry weather and hot summers, for example.

California does seem like paradise, with its delightful climate, ocean, mountains, etc. But paradise has attracted so many and the price to live there is high, with earthquakes, taxes and one-party government (I repeat myself) to offset its attractiveness.

I am skeptical of statistical analyses like those cited just because there are so many factors at work (such as whether you work or are retired). Of course, taxes make a difference, but the availability of jobs and a good climate may offset that. If you are mobile, why not move to a low-tax state? For example, Rush Limbaugh used to live and broadcast from NYC, paying NY/NYC taxes on his income. He was able to relocate to FL, moving his studio and saving himself millions annually. I don't think he is the only one. Rush left after NYS raised its so-called "millionaire's tax", after which Governor Patterson said, “If I knew that would be the result,” he said after a speech Thursday morning in Midtown, “I would’ve thought about the taxes earlier.” The governor had learned that taxes do matter.

A fellow I used to know pitched for the New York Yankees for one season in the 1990s. He and his wife decided, "Of course, we want to live in Manhattan if we're making $5 million, even if we have to pay 3 or 4% city income tax. It's New York."

Strikingly, however, he was the only one of 25 Yankees to live in New York City that year, according to their program.

I think a lot more celebrities live in NYC now than 20 years ago, probably due to crime coming down. Mayor Bloomberg made NYC much more friendly to the rich.

If you live in New Jersey like, say, Jay-Z and Beyonce but work in New York City, do you have to pay NYC income tax?

Here's a 2008 article on the most famous Yankee-about-town and his local income tax problems:

"Derek Jeter settled his case with tax officials, who had said the New York Yankees' captain should have paid three years of taxes as a New York state resident. A state official confirmed the settlement, which was first reported in Tuesday's New York Daily News. The official was not authorized to comment publicly and spoke to The Associated Press on condition of anonymity. Terms of the settlement were not disclosed. ...

"Tax officials contended Jeter should have been taxed as a state resident from 2001 to 2003. Jeter, who has a Manhattan apartment, said he was a Florida resident and didn't owe New York taxes. Florida does not have a state income tax; New York state and New York City do.

"Jeter's contract with the Yankees called for him to receive $11 million in 2001, $13 million in 2002 and $14 million in 2003. In addition, he has a $16 million signing bonus payable between February 2001 and June 2008. Jeter bought an apartment at Trump World Tower in October 2001, according to New York City real estate records."

California does seem like paradise, with its delightful climate, ocean, mountains, etc. But paradise has attracted so many and the price to live there is high, with earthquakes, taxes and one-party government (I repeat myself) to offset its attractiveness

OK earthquakes are no fun but in terms of as a reason to not live in CA or Washington this seems pretty weak. If you live in CA buy insurance to cover earthquakes, problem solved. Yes that's more cost than you'd get if you lived in NJ or NY but probably more people are likely to die each year getting into snow releated accidents in the Northeast than will die in CA earthquakes.

One party government? It wasn't that long ago that the GOP had the governorship of CA under Arnold. It isn't the Democrat's fault that was a failure and it doesn't seem obvious to me that Republicans are forever excluded from taking high office in either CA or NY. Unlike TX, both 'blue states' seem at least theoretically open to future Republican victories.

I think there's some confusion here about taxes. First, even high state taxes are much less than Federal taxes, so moving around to lower your tax bill can only have limited effect. Second, most people are tied to their jobs. The pitcher for the Yankees claiming he lives in Florida demonstrates that fact. Many people with high paying jobs are at jobs that are geographically tied. A Manhatten real estate broker or studio executive is not able to effectively do business out of a home in Kansas...even with the best internet hookup. A few people have massive incomes and can choose to live just about anywhere they want and it won't impact their ability to continue to earn income. For example, Stephen King can make millions and live anywhere he pleases.

One would think this class of people would be highly mobile, opting for the lowest tax places possible. Yet oddly many of these people don't seem to follow the low taxes. NYC and LA has a lot of celebrities who don't need to live anywhere near there.

I think celebrities is a more geographically restricted job description than you might think, since a large part of their job description is to be seen in the presence of other celebrities. I guess it doesn't matter *where* they're seen, but to do all have to be in one place.

I think you can make that case for some celebrities but not many, esp. not many of the big ones (and if the NY Yankees pitcher can argue he isn't a NY resident then so can just about anyone).

For example, I can see that Woody Allen probably has to live in NY. Does DeNiro? Does Tom Cruise really have to live anywhere in particular? Nicole Kidman I believe is a resident of Austrialia, doesn't really impact her ability to get a paycheck. So if you are at this level why live in CA or NY with its higher taxes rather than building a thousand acre compound for yourself out in Utah or Kansas?

Earthquakes are no fun? Speak for yourself. Earthquakes are lots of fun. I live about a mile east of the San Andreas fault, and when that section slips I'll get a big one. Bring it on!

This is one of the few places where workplace conversation can result in arguments over a tenth of a point on the Richter scale.

Did you feel that quake this morning? I'd say about 3.5.

No way! Probably less than 3. 3.2 at the most!

Wanna bet on that? We'll let the USGS website decide.

Biddle's paper isn't available yet for down-load (it's on Research Gate only), so it's impossible for me to comment on his methodology or conclusions. I did read his earlier paper on commercial air conditioning, which has the advantage of a comparison of firms (mostly restaurants and theaters) with air conditioning and without: customers could choose which to patronize without having to move across the country. The effect of residential air conditioning on migratory patterns, on the other hand, is far more difficult to quantify: there's BAC and AAC, and never the twain shall meet. Comparing migratory patterns and preferences BAC and AAC is comparable to religious preferences BCE (Before the Common Era) and CE (Common Era). Simple observation in my state, Florida, reflects vast differences in the volume of migrations AAC, in particular in the inland regions: the giant new cities that have evolved inland AAC (On Top of the World, The Villages, etc.) where ocean and gulf breezes don't exist but mosquitoes do. Of course, the Greatest Generation was the first that could afford to migrate to Florida after retirement, the first that could retire to an air conditioned home in Florida, the first that could retire to a relatively inexpensive air conditioned home in the inland regions of Florida; and when vast numbers of the Greatest Generation migrated to Florida, others followed in their wake to sell them goods and services.

Air conditioning made it all possible, but it's true that low "taxes" were and continue to be a significant contributor. What do I mean by low "taxes"? The cost of all those homes purchased by retirees and those who followed them did not include the cost of public goods (roads, utilities, schools, etc.) needed to accommodate them; those costs were spread among everyone, everyone except the developers. A few brave politicians suggested that developers pay "impact fees" to cover those costs, which the developer would pass on to the purchasers of homes. You'd have thought the politicians were proposing a progressive income tax in Florida! Of course, the absence of impact fees meant that ad valorem taxes and sales taxes would have to go up, way up, and they did. But not to worry: not so brave politicians came up with a novel approach for the millions of retirees already in Florida not to pay increases in ad valorem taxes (i.e., the annual increases were capped for existing home owners). Not only that, the cap is portable (it can be transferred from home to home if the retiree wished to buy another home). Low "taxes" mean that Florida's public goods are inferior and inadequate. Florida has, in effect, mortgaged its future (not by deficits, which the constitution prohibits, but inferior and inadequate public goods which affect today's quality of life but more importantly the future for those with inadequate and inferior educations). Not unlike the Bush tax cuts that were paid for with borrowed money.

I think you'll find that retirees tend to impose no new schooling costs. And as for capping property tax increases, retiree-owned homes tend to turn over relatively quickly (as they die off or, at least, become too old and frail to live independently). As the population of Florida has recently surpassed New York, it seems that the idea that Florida's 'cheep and cheerful' approach to taxes and public goods will be its eventual undoing is...well...not exactly supported by the evidence (which consists of many, many decades of rapid population growth).

Retiree's do require roads. They also have a habit of landing on Medicaid rolls after exhausting their assets or creative elder care lawyers position them as such. This can be a larger drain on a state's budget than school children can be.

But, unlike K12 schools, the federal government reimburses states for most of the cost of Medicaid. And when they are ready for the nursing home, many retirees will return to their original home states to be taken care of closer to their families. Although, despite the supposedly 'inferior services', it looks like in most cases, seniors would be better off staying in Florida for nursing home care:

If someone works in NY State and then retires to Florida, it makes NY look like a tax "payer" and Florida look like a tax "taker".

The new Florida resident will collect Social Security and Medicare in his new state. He will then spend that money in his new state, boosting the local economy without adding more competition to the labor market in Tampa.

It's a pretty good deal for Florida. They get a resident who spends down his savings while collecting large subsidies from the Federal government.

New York watches a lifetime of accumulated wealth slip across the border and get spent elsewhere.

Somehow the availability of air-conditioning has failed to turn the south into a steadily growing economic powerhouse...within the EU.

Quite right. But electricity is more expensive (relative to incomes), so A/C still isn't that common. The elderly have never had it so they don't know what they're missing; while the young head north for work.

The gap in productivity between Germany and Greece is FAR larger than the gap between Massachusetts and Florida.

Moreover, southern states in the US tend to be low regulation/less intrusive government areas.

Southern European countries tend to be highly regulated, very intrusive government areas.

Greece ranks #61 on the Ease of Doing Business Index (just between Tunisia and Russia). Germany ranks #14 (next to Canada).

Historically, almost all economists agreed with convergence where the poorer states -- mostly the old Confederacy -- would grow faster than the better off states. But since the late 1970s this convergence has not been as apparent. Maybe it is a function of the great stagnation.

Interestingly, when you compare state real per capita income of GDP the Northern great plains states seem to be the most prosperist.

BEA is now calculating regional price parities by state. According to them the richest 10 states by real income are:
(real per capita income as a % of national average)
North Dakota…139
South Dakota….118
New Jersey….110

and the poorest are:

West Virginia….91
South Carolina…..88
New Mexico…..86

I'm also surprised by Oregon an Utah.

Well, Oregon is more than Portland and Utah is more than Salt Lake City.

In wonder if they are using accurate CPI measurements for the Dakotas. From all the anecdotes I've heard, price of living has skyrocketed in all the areas where the high paynig jobs are.

Air conditioning isn't just a matter of personal comfort, it also affects productivity at work.

Although when it comes to outdoor comfort, cold can be dealt with by putting on more and better clothing whereas heat and humidity must simply be endured.

And high-cost places such as New York City can be relatively low-cost for those with low incomes. That is, high cost locations tend also to be high tax, but if you pay little in taxes and consume much in taxpayer-funded social services, such places may be a better financial deal for you.

It's hardly a secret that NYC's income distribution is bimodal, as the rich can afford to live where they wish (and to buy services that make it convenient to do so) while the poor are subsidized, but the middle gets forced out.

And the high earners are precisely those who are most able and most likely to leave a high-tax state for a low-tax state.

Anecdotally, it appears that a lot of later tech company employees (Facebook, Google, and smaller ones) who aren't likely to make eight figures from stock prefer Seattle to Silicon Valley because the cost of housing is lower and as noted above there is no income tax. The real earning / spending power for an average tech company employee is much higher in Seattle.

That has caused a lot of angst in the city, since it still restricts housing supply and consequently costs are rising. But overall costs are still much lower than SF, and both Seattle and outlying cities like Bellevue and Kirkland are building more than SF / Silicon Valley (which is admittedly not a high bar).

Anecdotally, it appears that a lot of later tech company employees (Facebook, Google, and smaller ones) who aren’t likely to make eight figures from stock prefer Seattle to Silicon Valley because the cost of housing is lower and as noted above there is no income tax. The real earning / spending power for an average tech company employee is much higher in Seattle.

But is it? Does 'real earning / spending power' mean something like "salary minus housing costs" as if all houses should cost the same price regardless of their location? I think even for most tech. workers, some geographical proximity with their employer is required. Most cannot simply live in Omaha and 'telecommute' to SF or Seattle.

So what is going on is:

1. High housing prices reflect actual high incomes. Jobs in 'high cost of living areas' do indeed pay more.

2. While allowing more housing units to be built would open an area up for more economic growth, it doesn't seem clear to me that it could necessarily ever 'create jobs' by lowering the cost of living. There's a lot more housing in Manhatten than SF. Yet all that additional housing just brought with it more jobs so as a result it still costs a lot to live there but to make a living there offers the opportunity for very high pay.

It seems to me that comparing state rates of growth in population and trying to determine their causes is a very tricky business. Among other things, the rates themselves are affected by the states' populations.

If a state with a population of 10 million has 5% of its residents move to one with a population of 1 million, and 5% of the smaller state moves to the bigger one, then the samll state grows by 45%, and the big one shrinks by 5.5%. What does that actually tell us about anything? Don't stocks matter as much as flows here?

"the high earners are precisely those who are most able and most likely to leave a high-tax state for a low-tax state."

I'm not sure I buy that. How easy it is for the NYC-based CEO or investment type to move to New Hampshire? Can the Chicago surgeon with a large well-established practice pick up and go to South Dakota?

The question is who can continue to be a high earner if they go to South Dakota? Stephan King, well it really doesn't matter where he writes his books. But as you point out, the surgeon, CEO, lawyer, etc. is not going to be able to earn as well in S. Dakota so it won't make sense for them to relocate unless their taxes and living expenses drop as much as their income will with the move.

This also casts doubt on the whole concept of states achieving higher economic growth via low taxes/regulation. Say Stephan King does move to South Dakota. His income will increase SD's GDP, that's cool. He will also add to demand in SD so that may create some jobs. But ultimately SD isn't really any better. SD 'itself' isn't producing anything new, its GDP bumps up simply because a publisher in NY send the direct deposits to another bank account...and unless you're a maid who gets hired to clean King's house, it isn't helping residents of SD all that much.

Comments for this post are closed