Deregulation and Airfares

Did deregulation really lower airfares?

On Twitter, Matt Stoller (who writes BIG on Substack) suggests that maybe airfares should be set by the government:

The airlines are making a great case for public regulation of the airlines. American Airlines is clear that a government agency should be setting prices and routes, because without that American Airlines cannot control its pricing and capacity.

…

Do I think the government should set prices and routes? I don't know how such a system would be administrable. It was how things worked until 1978 and it was a terrific system which saw dropping ticket prices and expanding capacity. What I do know is right now isn't workable.

He is referring to the high last minute fares that were still being charged by American, despite the desperate travelers stranded by the Southwest mess; in fairness, by the time he tweeted this, American and other major carriers had already capped fares, as they usually do in case of an emergency.

Stoller is actually echoing a standard airline talking point from the mid-1970s.¹ At the time, the government set airfares and determined capacity and routes, and deregulation was being pushed by Senator Ted Kennedy as a measure to introduce competition into the airline market and thus save consumers from being gouged. Airlines loved regulation, which protected them from new entrants (like Southwest) and price competition and pretty much guaranteed their profits and fat salaries. The airlines argued that regulation had given America the world’s best air transportation system and falling airfares, and that tinkering with it would jeopardize the quality of the system without providing any benefits.

Airfares have dropped significantly in the years since deregulation was signed into law by President Carter in 1978, but Stoller and the airlines are correct that airfares had been falling rapidly even before then. In fact, if you look at a chart, it’s hard to discern any difference in the trend in airfares before and after deregulation:

(Source: Airlines for America, Bureau of Transportation Statistics. I had to do some adjusting to compensate for methodological changes over time.) 

If airfares had been falling just as fast before deregulation, what was the point? Is there a way we can know if deregulation did anything to lower airfares at all? 

---

In the early days of post-war air travel, flying was expensive and slow. In the 1950s, airlines operated propeller planes like the Constellation or the DC-6, which carried around 60 passengers at a speed of 300 mph. A round trip transatlantic flight might cost $5,000 in current dollars; Pan Am famously offered installment loans for people who wished to travel overseas.

At the end of 1958, Pan Am began operating the Boeing 707, which ushered in the age of jet travel. A 707 carried about 140 passengers at close to 600 mph; in other words, it had twice the capacity and twice the speed of its predecessors. Jet travel was not only faster and more comfortable, it was much more cost efficient. Airlines overhauled their fleets, and by 1965, jets accounted for three-quarters of domestic travel, and the airline industry was booming.

The graph above indicates that airline operating costs per unit of capacity fell rapidly as jets took over in the early 1960s. It seems pretty clear that productivity gains in that era were mostly due to the ingenuity of the fine folks at Boeing, rather than anything the airlines or regulators did. Between 1958 and 1968, operating costs per available ton-mile fell 45% in real terms, despite rising union wages. 

Where did these efficiency gains go? Did they translate into lower airfares at the time? Well, yes and no. Government-regulated airfares were mostly steady, but revenue per passenger mile fell because of a mix shift toward cheaper coach and promotional seats. 

While revenue per passenger (i.e., airfare) was declining modestly, the graph above shows that revenue per unit of capacity was declining more significantly, usually not a good sign. One reason was more netural: airlines started to carry more freight, which carries a lower yield. The more worrying reason was that flights were taking off with more empty seats: load factor, or the percentage of seats that were filled with paying passengers, fell from 66% in 1951 to 53% in 1968. With regulated airfares, it was impossible to slash prices to fill seats, and emptier planes raised per-passenger costs. Low load factors would plague the industry into the 1970s, as airlines bought new planes they couldn’t fill.

Finally, as the graph suggests, efficiency gains were also translating into fatter airline profits. In 1965, airlines reported a collective operating profit margin of 14%. In 2022 dollars, they were turning a profit of 9 cents per passenger mile; a remarkable figure, considering airlines today only collect 16 cents of revenue per passenger mile, and are happy to make a penny or two per passenger mile, if they make anything at all.

The 1960s were an era of rapid advancement in aircraft design. The 727 entered service in 1964, the 737 launched in 1968, and the 747 came out in 1970. True, there have been some meaningful advances in fuel efficiency and labor efficiency since then, but the modern aircraft we fly today are still mostly descendants of these models. One would have expected that the 1960s and 1970s should have had the largest drops in airfares, since that’s when the biggest technological advances were implemented. 

This was the backdrop for the hearings on deregulation in 1975. Yes, airfares were falling a bit in real terms, but not nearly as much as might be expected given the abrupt shift from propeller planes to modern jet aircraft. The Civil Aviation Board (CAB) was responsible for overseeing the industry, but they were content to keep airfares high, and to allow the airlines to operate inefficiently. At the same time, airlines, not allowed to compete on price, competed on amenities instead. As Thomas Petzinger Jr. tells it in Hard Landing:

American installed a 64-key Wurlitzer piano in its jumbo jet lounges, with Frank Sinatra, Jr. at the keyboard on the inaugural flight. United staged “happenings” in its 747 lounges, featuring caricaturists, guitarists, and wine tastings.

…

Northeast Air proclaimed itself “the all-steak airline to Florida”. Delta introduced steak with a champagne accompaniment. National Airlines came back with free drinks of any kind…American wound up pouring so many Bloody Marys that it made millionaires of the people who bottled Mr & Mrs T Bloody Mary mix.

Ok, let’s take a look at the American in-flight 747 piano bar:

Now, there is nothing intrinsically bad about steak and a piano bar, but space at 30,000 feet is precious and one imagines that passengers might have liked the option to ditch the amenities and save a few hundred dollars. Alas, that was not an option with regulated airfares. This was the problem: price controls distort incentives, and so many of the efficiencies from new high-tech planes ended up going to empty seats and piano bars and free steaks instead of into the pockets of consumers.

---

In 1975, the Senate Judiciary Committee held hearings on airline deregulation, and produced a comprehensive report with data and predictions on how deregulation might play out. This gives us a set of hypotheses we can check against the actual evolution of the industry.

The general predictions were as follows:

1. In a competitive industry, airlines will choose to fly fuller planes and charge lower fares. Remember that fuller planes dramatically lower the realized cost per passenger, by spreading the fixed cost of operating the flight over more passengers. An airline that spends 16 cents per passenger mile at a load factor of 50% (that is, if it only fills half of its seats), will only be spending 10 cents per passenger mile with a load factor of 80%; in a competitive market, the savings should flow to the consumer. The same math applies if an airline crams more seats on each flight by reducing legroom. This was supposed to be the main channel through which deregulation would benefit consumers.

2. Airlines will cut back on amenities and compete on price. No more free steak, no more piano bars.

3. New entrants will drive the least efficient operators out of business, which will bring down costs overall.

4. Deregulation will have the biggest impact on airfares on longer flights. This is an artifact of the way regulated fares were calculated, which was strictly proportional to the length of the flight rather than the cost of the flight.

5. If deregulation ends up permitting selective discounting (as it did), then the lowest available fares will be much lower, compared to the existing system which permitted very little discounting.²

Most of these are pretty easy to confirm at a high level. The main prediction held true: even though the airline industry evolved in sometimes surprising ways (hub and spoke networks, loyalty programs), but load factor was indeed one of the biggest factors in lowering costs, rising from as low as 49% in 1971 to 84% today. All things being equal, one might expect such a sharp rise in load factor to lower per-passenger costs by 44% by itself (albeit with the unrealistic assumption that there are zero costs associated with additional passengers).

The other predictions mostly held up too. Airlines don’t offer free steak anymore. New entrants like Southwest and jetBlue have replaced failed incumbents like TWA and Eastern. Last minute airfares are still expensive, but leisure travel is now very cheap. Overall, there seems to be strong evidence that deregulation was the main factor in lowering airfares over the last half century.

We can drill down even further on the impact of deregulation by looking at some of the specific routes that the Senate studied back in 1975. First, they were very interested in routes that took off and landed in the same state, which happened to already be exempt from federal regulation. On routes like San Francisco to Los Angeles and Houston to Dallas, airlines were free to enter and set prices as they wished. 

They noticed that these unregulated intrastate routes carried a much higher load factor and had much lower fares (by about 40%) than comparable regulated routes that crossed state borders, just as theory would predict. As expected, the airlines had all kinds of excuses for why these routes might not be comparable, and the committee considered and dismissed them one by one.

We can revisit these routes today to see whether airfares on routes that had always been deregulated behaved differently from airfares on routes that were deregulated in 1978. We would expect that newly deregulated routes would see much bigger fare declines than routes that had always been deregulated.

(Source: 1975 Congressional Subcommittee Report, Department of Transportation. Note that the 1975 fare is the listed coach fare, while the 2019 fare is the average realized fare for all passengers, including first class, so it’s not precisely comparable.)

We can see that this is indeed the case. Intrastate routes (in italics) have actually gotten more expensive, adjusted for inflation, while interstate routes have gotten only a little bit cheaper. Note that we expect shorter routes like these to actually be the weakest test for deregulation, since they had the least inflated fares under the old system, and they skew more toward price-insensitive business travelers. There are a lot more substitutes for short routes (like driving), and short trips still face high fixed per-segment costs (e.g. airport costs).

With the same dataset, we can see that airfares fell much more sharply on longer flights. The coach airfare in 1974 for the most popular route in America, Los Angeles to New York, was nearly $2,000. The average realized airfare in 2019 of $722 on that route includes people who travel at peak times or at the last minute or opt for premium seating; a typical leisure traveler will pay $500 or less, and sometimes as little as $200. Almost everyone in 1974 had to pay the standard coach fare, while leisure travelers today can expect to pay 75%-90% less. Traffic on these routes is much higher today as well; we take it for granted today that we can fly whenever we want during the day, but high frequency was only made possible by the traffic stimulated by deregulated airfares.

For those that can’t live without the terrific amenities of the regulated era of air travel, take comfort that for the price of a transcontinental coach ticket in 1974, you can buy a lie-flat suite on the same route today, fine dining and open bar included. It’s enough to make the modern system seem workable.

Further reading: We looked at the Great Texas Airline War of the 1970s a few weeks ago, a more relevant story today.

1.  It’s fine to be skeptical of businesses, or of any institution for that matter. The problem with leaning too heavily on skepticism of business as a guide to policy is that many if not most policies benefit one set of businesses while hurting another; there is no “anti-business” stance available allows one to oppose all businesses. For example, airline deregulation benefited new entrants like Southwest, while crippling weaker incumbents like TWA; continuing regulation would have done the opposite. Reflexively taking the anti-business position without analyzing the issue in full will inevitably cause one to accidentally stumble into some very pro-business / anti-consumer/labor stances, when the presumed intent is to take out with pro-consumer/labor positions. We saw this with housing: supposedly progressive NIMBYs (who see themselves as opposing developers) end up on the same side as major corporate landlords like Blackstone, who also want to see new competition blocked so they can raise rents. (Just to cover both sides, conservatives do the same thing, except with government instead of business.)

2. Keep in mind that “yield management”, the modern practice of charging wildly different fares to different passengers on the same flight based on when they bought their ticket, wasn’t adopted by airlines until the mid-1980s. Back then, discounting was much more limited, with different fares for flights at off-peak times and families. Yield management is responsible for some of the deeply discounted fares that leisure travelers enjoy today, and probably would not be possible in a framework of regulated airfares.