Oil, Putin and windfalls

11 Mar 2014

John Aziz argues that unleashing the US Strategic Petroleum Reserve won’t hurt Russia. The premise is a proposal to release oil from the reserve to lower oil prices:

As of now, Putin is profiting from his invasion. That is because oil prices are up on the risk of a supply disruption. This enriches the Russian state budget, half of which is supported from oil and gas exports. But economist Philip Verleger notes that prices can go down as well as up, and he recommends inflicting pain by engineering the former.

John makes some good points. I found this one particularly interesting:

Second, even if the strategy lowers oil prices, it will injure large American and European companies, as well as their employees and customers. The revenues of oil giants like Shell, Exxon Mobil, and BP — each of which employs thousands of people — are dependent on the price of oil. The costs of a global price drop will either be felt in falling profits, or passed onto consumers.

Presumably the plan is to release just enough crude oil to lower prices to the level they would be at were it not for the Ukrainian crisis, and perhaps even stopping short of that. The oil industry would lose a windfall that they only gained because of a political crisis. Maybe you believe that windfalls should be protected at all costs, but I think there’s a case to be made that confiscating this windfall is an acceptable cost if the rest of the case for releasing reserves is solid.

He goes on to propose freer natural gas exports instead:

As I have argued before, natural gas is a far more appropriate tool to gradually undermine Putin's leverage over Europe. The U.S. has world-leading natural gas reserves, and boasts very cheap natural gas prices compared to Europe. Many countries in Europe are totally dependent on (expensive) Russian natural gas that Putin can cut off whenever it is politically convenient.

If unrestricted natural gas exports are allowed, natural gas prices in the US will probably rise, with the consumers paying much of the cost. Why is it okay to take away US natural gas consumers’ windfall (caused by a government policy of restricting exports), but not oil company shareholders’ and employees’ windfall (caused by a government policy of restricting the petrolum reserve)?

Is there a gender gap in tech salaries?

03 Mar 2014

Cynthia Than at Quartz brings us the delightful headline “There is no gender gap in tech salaries” based on a study from AAUW. The study is based on non-public data and is, unfortunately, not very rigorously documented.

The key conclusion from the article is that there is no gender gap in salaries for engineers and those working in math, computer and physical science occupations. As far as I can tell, that result comes from this figure in the study:

Figure 8

From this figure, you would think that male and female engineers have exactly the same average earnings, $55,046. There are a number of problems with the way this is presented and with the underlying study and its interpretation.

If you go to the underlying tables, you’ll find that male and female engineers do not have the same earnings. This data is based on earnings for full time employees in 2009, a year after the sample graduated college in 2008, and is restricted to those who were 35 years or younger at graduation. I wasn’t able to recreate the figure precisely using the Department of Education’s online tool, but I got pretty close to overall average earnings for this group of engineers of $55,076. Here’s what happens when you chart it out by sex:

Gap for Engineers

That’s right: in levels, female engineers make about 11% less than men one year out of college.

Both Cynthia Than and the researchers behind the study claim that the difference is not statistically significant. That may be true, but it’s still misleading to simply show the same bars for the two groups. If you want to make a point about significance, use error boxes or something like that.

Is there, in fact, an insignificant difference? I’m not sure. NCES has a helpful tool for calculating the t-statistic:

The tool tells us that the value is 2.25, which means that the gender gap is statistically significant. I am not sure because it’s possible that more thought was put into Figure 8 than is apparent from the notes. Let’s get back to that.

You might think that it’s weird to only look at gender gap for earnings one year after graduation. It could be because the data set only contains that variable. But why did they pick that particular data set when other datasets, such as the American Community Survey, have more data? Perhaps to control for additional variables such as the rank of the college and GPA that are not present in ACS. There’s also an argument that restricting to earnings one year out of college helps control for the motherhood penalty and other factors.

That could be true, but it still only tells you something about earnings for a very limited set of workers. You wouldn’t be able to conclude that there is no gender gap in tech salaries; for example, they could show up later in the career.

Figure 13 of the study has what appears to be a carefully done regression that shows a statistically significant gender gap of 6.6% across all occupations, controlling for things such as hours worked, economic sector, undergraduate GPA and whether the undergraduate degree was from a very selective institution. That’s the only specification shown, so we are not shown any evidence that these extra controls actually affect the size of the gender gap (the regression coefficient on female).

The conclusion of the Quartz article was that there is no gender gap in tech. It’s not clear that Figure 8 is based on a regression that controls for anything at all. The notes say:

This chart shows average earnings 2007–08 bachelor’s degree recipients employed full time in 2009 and excludes graduates older than age 35 at bachelor’s degree complation. In occupations with red and green columns shown, men earned significantly more than women. In occupations with one blue column shown, there were no significant gender differences in earnings one year after graduation.

There’s no mention of control variables, or a regression, or how the regression was made (for example, restricting to tech workers, or introducing gender×occupation interaction dummies). It’s still a mystery how they arrived at the conclusion that there’s no difference, or why they created a misleading chart, or whether they performed a full analysis of the gender gap by occupation, and if so, why the results are not reported. And if we don’t control for these things, what’s the point of using a small study that only has data for earnings one year after graduation?

I do think it’s safe to say that this study, as reported, does not have enough evidence to conclude anything about occupational gender gaps, whether it’s for earnings one year after graduation or at any other time in workers’ careers.

Another question is whether all of those controls are appropriate (if they were included). One of my teachers once told me that social scientists overcontrol. What he meant was that a social scientist tends to be interested in the partial effect of one variable controlling for every other conceivable variable, but the public or policy makers may not be interested in the partial effect: if women make less than men, they may think that’s a problem, whatever the reason. You may still want to include plenty of controls for efficiency, but your reader may be more interested in a result that has those controls integrated away. In this case neither undercontrolled nor overcontrolled results are accurately reported.

Update: Cynthia Than tells me that the significance results from Figure 8 control for education, work status, occupation and career timing. As far as I can tell she got that in private email communication with one of the study’s authors. (The authors’ email addresses are not in the report or readily accessible on the AAUW website, by the way, so there’s no way for anyone else to ask questions about it.)

I don’t want to belabor this point too much because even if everything is done correctly, Quartz is still trying to draw general conclusions about the gender wage gap based only on salaries one year out of graduation. This is a point about external validity: all the arguments in favor of using salaries one year after graduation only establish that those salaries can be measured with less error and fewer confounding factors than, say, salaries 20 years after graduation. They do not establish that salaries one year after graduation are of particular interest.

However, I still think we should be careful with a report that lacks full documentation on its methodology and how the results are constructed. We are not given regression specifications, results from alternative specifications, sample sizes, t-statistics, or even a full list of control variables for Figure 8 (except what was privately communicated to Than). I don’t necessarily fault the study authors for omitting this in a report for general consumption when Figure 8 is only a minor aspect of the whole report. But without more information about that aspect of the study, I would be very cautious in relying on it, and in making the general conclusion that there is no gender gap in tech salaries.

I’ve been trying hard to find good data on this and again it’s been hard. The NCES tool initially looked useful. Presumably because of sample size related privacy issues, it refused to give me average wages by gender for tech professions other than Engineers. That raises concerns about statistical power. It’s possible that the online tool is based on a smaller sample than what the AAUW researchers had access to. If only either the online tool or the AAUW study would report sample sizes!

Local loop unbundling

24 Feb 2014

Local loop unbundling is common in Europe and is a way to ensure competition in the retail ISP market. It’s also a way to sidestep many of the issues surrounding net neutrality: if one ISP does something distasteful, you can just switch to another one.

The basic idea behind LLU is that you don’t have to buy Internet access from the same company that owns the wire (copper pair designed for phones or coaxial cable designed for television). For example, I might get Internet access through an independent ISP. They would then pay a monthly fee to Time Warner Cable, which owns the physical cable, and have some equipment in a TWC facility. My ISP would have a backbone that includes a connection to the TWC facility. If I am unsatisfied with my ISP, I could switch to a different one that services my neighborhood and has a deal with TWC.

TWC still has a monopoly on access to the cable (the local loop), so their rates might have to be regulated. This regulation is difficult in some countries where the local loop owner also operates a retail ISP business, because TWC might price their own Internet access offering too low compared to competitors because their own cost is much lower. You then need to come up with some scheme to come up with and mandate the “right” price for the local loop. An alternative could be to ban TWC from offering retail Internet access, and let them charge ISPs whatever they want.

If you think that monopoly rents are too low in an LLU scenario, you could set wholesale loop rates at a very high level, or even not regulate them at all (and ban the provider from providing retail services).

LLU can operate at different levels in the network. One method is to have the physical wire from my home terminate in equipment owned by my ISP. If I change ISPs, the other end of the wire would have to be physically or electrically switched to a different ISP’s equipment. That can become impractical as technology changes; for example, with AT&T U-verse, your wire terminates in a box on the street less than a few blocks away, and it’s not possible for lots of ISPs to all be present in every box. Instead, you would do the unbundling at different levels; bitstream access is a common way to do it for DSL type connections.

Mandating LLU for cable Internet access in the US would be a huge step and it’s not clear that the FCC has or wants the authority. It might also seem like an overreaction to what is, in the short term, a relatively minor problem. But if LLU had been implemented in the US a decade or more ago, we probably wouldn’t be in this mess.

The concept of LLU doesn’t just apply to telecommunications. In the US, FedEx and UPS operate shipping services where they move your package most of its way across the country, but then hand off to USPS for the last mile. SmartPost and MI happen to be among the crappiest shipping methods known to man, but that’s only weakly related to the fact that they hand off to USPS.

The motives of cable companies

24 Feb 2014

Tyler Cowen:

In a nutshell, the gatekeeper won’t want to exclude the programs which consumers really want.

Kevin Drum:

Comcast wants us to use the internet only in ways that don’t interfere with the money they make from bringing TV and other video streams into our homes.

Why Netflix paying ISPs is bad

23 Feb 2014

A few days ago we received the apparent good news that Comcast and Netflix are now exchanging traffic directly. Today, the Wall Street Journal reports that Netflix is paying Comcast for peering. This sets a disastrous precedent for the future of the Internet.

I’ve seen several attempts to explain why it’s bad that Netflix pays for Internet access. Some of them have been a bit unsatisfying. In one analogy I saw on Twitter that appears to have been withdrawn, it was compared to UPS opening up packages and delaying them based on the contents. But that’s not quite right: UPS already charges shippers extra for faster shipping, and they certainly won’t ship anything for free.

In the Netflix–Comcast deal, Netflix and Comcast entered into a peering arrangement. The way this works for high bandwidth content providers, Netflix is basically shipping streaming bits directly to the customer’s doorstep at multiple locations throughout the United States. A Comcast customer who watches House of Cards in Philadelphia will receive the stream from Netflix through a Comcast facility somewhere in the Northeast. (Comcast could, reasonably, require the stream to be delivered to a facility in Philadelphia.) Netflix has already paid to bring the video from their servers all the way to the nearest Comcast peering point.

The analogy I would use is that your apartment building charges Netflix to bring their red envelopes through the door. This is unfair because you’ve already paid for your mailbox and lobby (and maybe even a doorman) through your rent. Netflix should pay to bring the DVD to your doorstep, and they do. (I won’t consider the special case where USPS is your landlord.)

Your building should not get to charge shippers and delivery people extra: you don’t really have a choice of an alternative mailbox/lobby provider if you think their terms are unfair. And you already paid for this service.

You could always move to a different building. But guess what: every building in the neighborhood (or if you’re in a smaller town, the whole town) is owned by the same company.

Getting back to my earlier post on net neutrality, the problem isn’t so much that Netflix has to pay a little extra. It’ll add a bit to their costs, in the longer run they have to charge you a bit more, but a Netflix subscription is already cheap. The problem is that we lose a culture of permissionless innovation:

Anyone can buy Facebook ads (maybe BuzzFeed gets a better deal?), but the kind of access that Netflix can get by paying is only available to big, established companies who can get ISPs on the phone and get them to take meetings. Or who have the wherewithal to meet with every major residential ISP in the country (or the world, if they want to be global).

And who knows if Netflix could ever have become a big, established content provider if established players had been allowed to veto their growth at every stage. (This would be a lot more tolerable if the cable companies got out of the content business and become pure ISPs.)

There’s a growing trend of mood affiliation in favor of high and increasing monopoly rents in as many markets as possible. It’s true that in contestable markets, monopoly rents can provide an incentive to invest and to actually contest the market.

However, I see very few attempts to ask “how much?” and “how?” There are potential benefits to the existence of rents. There are costs too, in the form of reduced consumer surplus (you have to pay more), less innovation, and in the case of the Internet, the loss of permissionless innovation. Some methods of collecting rents are more harmful than others.

In the US, patent terms are 20 years from priority date; why not make it 30 years, or 40 years, or 120 years? I think the reason is obvious, but nobody ever thinks this way about ISP rents. Until the Netflix–Comcast deal, a limiting principle for rents accruing to an ISP was “you can screw the customer as much as you want, but you can’t screw the content provider”. No more.

Update: Changed the title to reflect the fact that this post is not written from Netflix’s point of view. Also, I should add that my description of peering reflects what Netflix is now paying Comcast to do, and what they do with other providers such as Cablevision on a settlement free basis, but does not describe how peering generally works. In particular, what I describe is the opposite of the “hot potato” routing in most peering agreements.