What makes a soda tax regressive?

I’ve been running in to articles about soda taxes, after Philadelphia’s three-cents-per-ounce soda tax became an issue in the Pennsylvania Democratic primary. A strange omission I’ve been noticing is that some pundits can’t quite seem to name exactly what makes a soda tax regressive.

Jonathan Chait mostly evades the issue, until claiming that Bernie Sanders’ opposition to a soda tax on the grounds that it is regressive is inconsistent with Bernie Sanders’ own tax plan:

Indeed, it is very hard to understand how Sanders’s opposition to it can be reconciled with his own platform, which — both in its specifics and its general theme — rests upon higher taxes on the non-rich. To oppose a new tax-and-transfer plan based solely on the regressive character of the financing source, without any consideration of the benefits of the spending, would rule out his own plans as well.

Chait seems to try to muddy the argument a little by arguing that a regressive tax isn’t regressive if it has progressive effects, and it is by this virtue that Sanders’ tax platform and the Philadelphia soda tax are progressive, but nestled in this argument is the implication that the Sanders tax plan is overall a regressive one. It’s hard to see how it is. A regressive tax is one that taxes a greater percentage of a poor person’s income than a rich person’s income, but the Sanders tax plan increases the tax burden of the lowest quintile of Americans by 1.2% while increasing it on the highest quintile by 12.8%.

This New York Times article gets closer to being able to explain why a soda taxes are regressive:

It can be seen as achieving an admirable public health goal of less sugar consumption or as a very regressive tax that falls more on the poor than the rich, since the poor tend to drink more soda.

It’s certainly true that the fact that poor Americans drink more soda makes a soda tax more regressive, but it glosses over the fact that consumption taxes are by nature regressive. If every American drank a single can of soda every day, they’d all pay the same dollar amount in tax, which means people with less income pay a higher portion of their income in soda tax. Rich people could drink several times more soda than poor people and a tax on soda could still be regressive, as long as the portion of their income they spent on soda (and the soda tax) was smaller.

This might warrant a separate post, but I do want to revisit the idea that a tax isn’t really regressive if it is specifically levied to fund a program that helps poor people more than they are taxed. It seems to me that revenue sources are fungible: any money going from a soda tax to a universal preschool program could be replaced by the same amount of money coming from, say, property taxes, while the soda tax revenue goes toward maintaining historic buildings in wealthy neighborhoods. Even if Philadelphia were to specifically tie the revenue from taxing soda to funding its preschool program (as I suspect it has), what if the soda tax works as intended and people drink significantly less soda? There’s probably some limit to how much you can tax soda before you reach the peak of the sin-tax Laffer Curve and people start smuggling their Mountain Dew in from Delaware. When that happens, do you find a new source of revenue (proving that the revenue was fungible all along) or leave the preschool program underfunded? I’m sure there are practical political reasons for tying a certain source of revenue to a certain program, but it’s hard to see how a progressive program can make a regressive tax into a progressive one by association.

A quick look at US trade

I’m working on a post about the Autor et al paper from earlier this year regarding the impact of trade with China on US manufacturing, and in doing so I’ve been playing around with some trade data from the US Census Bureau. Here I’d just like to put up some of what I’ve found.

First, just the basics:


The country specific data that I’m using only go back to 1987, unfortunately. Looking at trade overall we see that while the US has consistently run a deficit, it began to balloon  around 1997 until 2006 or so. After the global recession (which really had a big impact on trade!) the deficit has remained stable at a level slightly less than the previous peak.

Looking at the evolution of our trade partners over time, here’s what happened to our top five trade partners plus China from 1987 until today.



Here we can see the remarkable rise of China in context, and it really is remarkable. Note the acceleration after China joined the WTO at the end of 2001. It’s hard to be sure yet, but it also looks like the curve reached another inflection point around 2006 or 07, and since then Chinese imports have been increasing at a decreasing rate. Also apparent is the steady rise of trade with Mexico, which also looks to have taken off after NAFTA in 1994. Taiwan’s imports have basically remained steady the past 30 years, as have Japan’s.


Looking in terms of percent of total imports, the patterns of Chinese and Japanese trade are almost exact inverses of one another. Just before the beginning of Japan’s first lost decade, they provided nearly the same share of US imports as China does today. I was just a toddler during the 80s, but from what I understand there was a real fear that Japan was going to overtake our economy, much as people fear China’s rise today. I tend not to worry too much about China overtaking us economically for a number of reasons, and seeing Japan stall out and stagnate for essentially 20 years is one of them. On the other hand, China does have an order of magnitude greater population, and they’ve got quite a ways to go before getting to our level of per capita GDP. I’m sure Mike has a word or two about that.

Finally, I wanted to see if the nature of trade has changed over time. In industrial organization, often something called the Herfindahl Index, or HHI, is used in measuring the level of concentration in an industry. It ranges from 0 to 1:  values closer to 0 indicates a large number of small firms, and values closer to 1 indicate a small number of large firms. So higher levels suggest a decrease in competition. I constructed an HHI type index for all of the countries that trade with the US using their ‘market share’ of total US imports and exports for each year in the dataset.


Again, higher values indicate higher concentration: more trade with fewer countries. Now, overall the range of values is pretty low and pretty similar across imports and exports. As a final quick check, I wanted to see if these variations were simply due to the value of the dollar relative to other currencies. When the dollar is strong, our exports are less competitive and so the trade HHI should be higher in those times due to reduced trade, and vice versa for imports.

As quick regression I ran

hhi = year + twdi + nafta + wto

where hhi is the trade HHI *100^2 (basically to make things easier to see), twdi is a Trade Weighted Dollar Index, and nafta and wto are dummy variables for the years in which NAFTA and China’s membership in the WTO are in effect.



lm(formula = im ~ Year + twdi + nafta + wto, data = twdi)

 Min 1Q Median 3Q Max 
-61.324 -22.092 -9.477 22.190 98.867 

             Estimate Std. Error t value Pr(>|t|) 
(Intercept) -7356.091   5114.772  -1.438 0.16329 
Year            4.212      2.550   1.652 0.11161 
twdi           -1.473      1.159  -1.271 0.21579 
nafta         -89.175     30.399  -2.933 0.00726 **
wto           -70.701     32.236  -2.193 0.03821 * 
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 40.82 on 24 degrees of freedom
Multiple R-squared: 0.5255, Adjusted R-squared: 0.4464 
F-statistic: 6.644 on 4 and 24 DF, p-value: 0.0009525



lm(formula = ex ~ Year + twdi + nafta + wto, data = twdi)
 Min 1Q Median 3Q Max 
-77.265 -20.362 -3.355 15.429 53.024
              Estimate Std. Error t value Pr(>|t|) 
(Intercept) 10419.4969  4071.2267   2.559 0.0172 * 
Year           -5.1497     2.0296  -2.537 0.0181 * 
twdi            6.7052     0.9224   7.269 1.65e-07 ***
nafta          63.1336    24.1972   2.609 0.0154 * 
wto            21.9839    25.6586   0.857 0.4000 
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 32.49 on 24 degrees of freedom
Multiple R-squared: 0.8132, Adjusted R-squared: 0.7821 
F-statistic: 26.13 on 4 and 24 DF, p-value: 1.936e-08

As expected, exports are highly correlated to the TWDI, and NAFTA seems to significant to both imports and exports, but since this is time series data for a relatively short period of time, I’d take that with a grain of salt. It’s interesting that the TWDI doesn’t seem to affect the import index; I’ll have to think of why that would be.

That’s all for now, though I could spend days poring over this. What I really wish we had was more historical trade data, as pretty much the entire time period covered by this Census Bureau data is from the era of liberalized trade. To even go back another couple decades would be really interesting.

Some thoughts on the US military-civilian relationship

I’ve been reading former Secretary of Defense Robert Gates’ 2014 memoir, Duty, and it’s been fascinating. Probably the most interesting part so far is Chapter 10, which focuses on the war in Afghanistan. I don’t think it will surprise anybody that in reading this book it becomes clear that there is a strong disconnect between the military and the Democratic party. This is on full display during the decision process of whether to increase troop levels in Afghanistan early in the Obama presidency.

At this point in the story, Obama has been in office less than a year and Stanley McChrystal has recently been named senior commander in Afghanistan. As one of his first actions, McChrystal ordered a complete assessment of the situation on the ground. This assessment concluded that an additional 40,000 troops were required, in addition to the 21,000 more Obama had recently authorized just months earlier. Understandably, Obama was reluctant to get drawn further into Afghanistan, given that he was elected partly on the promise to disentangle the US from its wars.

“On September 13, the president chaired the first of nine— by my count— very long (two-to-three-hour) meetings on McChrystal’s assessment and Afghan strategy. Two days later the Senate Armed Services Committee held a confirmation hearing for Mike Mullen’s second term as chairman of the Joint Chiefs of Staff, at which time he forcefully argued for more troops in Afghanistan. He was implicitly critical of the vice president’s views, saying we could not defeat al Qaeda and prevent Afghanistan from becoming a safe haven again “from offshore.… You have to be there, where the people are when they need you there, and until they can provide for their own security.” The president— and everyone else in the White House —was livid, seeing the testimony as another effort by Mullen and the military to force the commander in chief’s hand. [“Duty: Memoirs of a Secretary at War” by Robert M Gates p. 366]


“Then the biggest shoe of all dropped . On Monday, September 21, The Washington Post published a detailed story by Bob Woodward on McChrystal’s assessment, clearly based on a leaked copy. The four-column-wide headline read “McChrystal: More Forces or ‘Mission Failure.’” The Post had given us advance warning it was going to run the story, and over the weekend Cartwright, Flournoy, and Geoff Morrell negotiated with Woodward and others from the Post to remove sensitive numbers, references to intelligence gaps, Special Forces unit designations, and the like. They had some success, but they could not redact the political bombshell the story represented. The story ended with a quote from the assessment: “Failure to provide adequate resources also risks a longer conflict, greater casualties, higher overall costs, and ultimately, a critical loss of political support. Any of these risks, in turn, are likely to result in mission failure.” After I left office, I was chagrined to hear from an insider I trust that McChrystal’s staff had leaked the assessment out of impatience with both the Pentagon and the White House. If so, I’d be very surprised if Stan knew about it. [p. 367]


“An infuriated president, Mullen, and I repeatedly discussed what he regarded as military pressure on him. On September 16 , Obama asked us why all this was being discussed in public. “Is it a lack of respect for me? Are they [he meant Petraeus, McChrystal, and Mullen] trying to box me in? I’ve tried to create an environment where all points of view can be expressed and have a robust debate. I’m prepared to devote any amount of time to it— however many hours or days. What is wrong? Is it the process? Are they suspicious of my politics ? Do they resent that I never served in the military? Do they think because I’m young that I don’t see what they’re doing?” Mike assured him there was no lack of respect. I said we just needed to shut everyone down until the process was complete.” [p.368]


“Again and again I tried to persuade Obama that there was no plan, no coordinated effort by the three military men to jam him. I said that if there had been a strategy to do that, they sure as hell wouldn’t have been so obvious. I reminded him that McChrystal had never had a job before with the kind of public exposure he now had, that he was inexperienced and a bit naïve about dealing with the press and politics. I said Mullen and Petraeus were both on his team and wanted to serve him well; but particularly when testifying, or even when talking to reporters, both felt ethically compelled to say exactly what they thought, however politically awkward. I told the president that Mike’s independence had annoyed Bush as well. My assurances fell pretty much on deaf ears, which I found enormously frustrating and discouraging.[p. 369]


Why this mutual level of distrust between Democrats and the military? Note that this is not unique to the Obama era: the Clinton administration also had several high profile disagreements with the military. I can think of two, non mutually exclusive explanations. One is that Democrats simply haven’t had that much experience leading the military in the post Vietnam era. When Obama took office, Democrats had controlled the executive branch in only 12 of the past 40 years, and so there wasn’t a deep bench of people with defense or military experience. Under this hypothesis, it’s then unsurprising that Obama appointed two Republicans to be Secretary of Defense.

The other potential explanation is a cultural one. In our current political alignment, liberals are associated with the Democratic Party and conservatives with the Republicans. No surprises there. Furthermore, the military tends to be more culturally conservative: the values of patriotism, honor, obedience to authority, etc. are important across both members of the military and those who identify as Republican. With the high levels of polarization present in our politics, it should not be surprising that levels of trust across ideological groups is low, and hence Republicans tend to have have better military bonafides. This is evidenced by polling, which shows that members of the military strongly identify as Republican and conservative (but as an interesting aside, the number of Republicans has dropped precipitously over the past decade, in favor or more independents and libertarians).

Looming larger than either of these possibilities is the fact that there is a broader disconnect between the military and civilian society that has been growing for decades. According to the Pew Research Center,  the percentage of veterans in Congress is at record lows, as well as the fraction of the civilian population that has served. The divide was visible even 20 years ago, as evidenced by this Thomas E. Ricks piece from 1997 which also highlights an increasing conservatism and politicization of the military officer corps, trends which I imagine only intensified during the 2000s. This is a big problem that deserves greater attention. When the more hawkish elements of our political establishment want to engage in military solutions to problems in the world, an unfamiliarity with military capabilities can lead to unrealistic expectations (see the last 15 years). And the more dovish, with their mistrust of the military, will be reluctant to intervene in the instances where it is required. This leads me sometimes to believe that some sort of required national service would be healthy for the country, though that’s a topic I’ll leave for another day.
Now perhaps a more interesting question is, must it be the case that a military is culturally conservative? After all, Starfleet seems pretty liberal ;p. But seriously, it could be the case that the conservatism of the military is a compositional effect: that is, if for whatever reason the people who join and remain in the military tend to be more conservative, then this would bubble up into the leadership and so on. The all volunteer armed forces would of course only intensify this self selection process. On the other hand, it could be the very nature of military operations that enforces a sort of conservative culture; any organization that requires high levels of coordination and trust to accomplish its goals will need to enforce a homogeneity of values and inculcate adherence to authority, etc. This is also a topic I’d like to explore in the future, because the question of ‘what would a liberal military look like?’ isn’t often addressed.

I’d like to close by acknowledging that this is all still at the speculative phase, and I haven’t defined my terms well or presented much data to defend my claims. Additionally, there are multiple dimensions to liberal and conservative values and I’ve been grouping them together quite broadly. This is a topic I’ve only begun to think about recently, and so I haven’t yet delved into the relevant literature. 

Particular Contradictions

A list of contradictions within the Chinese government:

  • China is a country run by a Communist Party that practices free market capitalism. The reality isn’t quite as simple as that, there are large parts of the Chinese economy that are still controlled by the state, and Communist ideology is very much a part of the education system (the college students I taught were all required to take a class on Mao Zedong Thought). The Party still presents itself as Communist (I have trouble imagining how it could walk that back), but seems to get quite a bit of mileage out of the elements of Marxist and Maoist thought that were less straightforwardly hostile to market capitalism. I imagine they could maintain this fiction for a long time to come, but Communist ideology doesn’t seem to be the active force of legitimacy for the CCP that it once was.
  • China’s apparently content-empty ideology has led the CCP to rely on two sources of legitimacy: economic growth and nationalism. As the economy slows down, it will have to increasingly rely on nationalist sentiment, which often means actively contesting its territorial claims. The Diaoyu/Senkaku Islands dispute is a personally-felt issue for many Chinese; it isn’t unusual to find signs proclaiming “The Diaoyu Islands belong to China!” when you are going about your business. I used to own a mug with that very slogan. China has strategic reasons to maintain claims on gas-rich maritime territory, but backing away from those claims for diplomatic reasons would now be very difficult. In the meantime, the claims that China would undergo a peaceful rise ring increasingly hollow, and China’s neighbors, particularly in Southeast Asia, are aligning themselves to contain Chinese ambitions
  • To the degree that the CCP does still rely on economic growth as a source of legitimacy, it does so by projecting an air of competence and practicality. The Party, however, is only so tolerant of internal dissent (apparently especially so under Xi Jinping), which artificially limits the solutions available for consideration.
  • It’s no accident that Xi Jinping followed Hu Jintao, a leader so bland he read from index cards while meeting foreign heads of state. China has been looking for a leader to match its increased stature, someone who would clean up corruption, manage the slowing economy, and begin actively pushing China’s foreign interests. These are tasks that lend to the decisiveness of a single leader, but the Communist Party has been deeply skeptical of allowing a single individual to accumulate too much individual influence after Mao decided to use his own influence to repeatedly purge the party. It’s hard to know who to root for in the growing fight between Xi and the party members grumbling about him. (The recent policies of restricting VPNs and stepping up restrictions on foreign journalists seem to have come from Xi, so I am inclined to root against him.)
  • Mao Zedong, a man whose collectivization and industrialization policies were responsible for the deaths of tens of millions of people, is still used heavily in official iconography. He is on every bill from the 1 yuan note to the 100 yuan note (the two or three smaller notes have ethnic minorities on them). Mao seems to be revered and despised in equal measure. Committing to the veneration of Mao, even considering the popular “Mao was 70% right, 30% wrong” line, means the party is committing to restricting information about Mao’s worst excesses.
  • I once ran into a quotation by Deng Xiaoping that I can no longer track down, but it went something like this: Corruption will be the end of the party, but without corruption, there would be no party. Real quotation or not, it still hits at an ongoing issue: If you don’t allow low- and mid-tier Chinese officials to expense fancy dinners to the party, what’s the point of carrying out the will of the Central Committee?
  • The CCP has maintains the illusion that the leaders don’t enrich themselves and their families through their positions in the party. It’s pretty clear that this is untrue, and Xi Jinping’s enduring popularity in China is linked to the fact that he made fighting corruption a major aspect of his administration. The thing is, Xi’s family is clearly quite wealthy. (Reporting on the wealth of CCP elites is such a surefire way to not get your press credentials renewed that Bloomberg notoriously pulled an article about Xi Jinping’s family. The New York Times remains blocked because of a story they published on then-Premier Wen Jiabao’s family.)
  • Allowing foreign journalists to publish what they please means more articles revealing hypocrisy among the CCP elite, which undermines the authority of a party which has used the phrase “Serve the People” as a central slogan for decades. Tightening restrictions on foreign journalists tends to piss off said journalists, which is a really good way to ensure hostile coverage. You don’t need to be in China to report on a stumbling economy and military adventurism.
  • China is increasingly suspicious of all aspects of the liberal international which do not provide special allowances for China. The Asian Infrastructure Investment Bank seems to have been a practice in creating an international organization that gives China the pride of place usually reserved for the United States or the OECD countries as a whole. They seem to have invented their own definition for ‘freedom of navigation’. If you buy that the network of international organizations and agreements are a major part of what underpin international trade and allowed for globalization of the world economy, then China may well be tugging apart the threads of the global order that allowed for its sudden rise as an economic power.


Clientelism and Political Parties


Imagine a country with a clientelistic political system composed of a series of patronage networks corresponding to political parties, lots of poverty, elections, weak democratic institutions, and low administrative capacity.

The leaders of the patronage networks compete for the spoils of the central government, which they then push down the network. In practice, at the top this looks like envelopes full of cash and juicy contracts handed out to deputies, at the bottom it looks like local leaders handing out bags of grain and minor jobs to poor people (who make up a good chunk of the society). Each node in the network skims some fat off the top. In exchange, each layer swears political to fealty to higher layers in the network, through votes, loyalty, and a willingness to get out into the streets when called upon.

In this country, political parties do not trade in ideology, but rather the exchange of political power for resources. They frequently change names, symbols, and rhetoric. These parties lack strong, formal organizations and instead rely on strong informal relationships.

This appears to be a glass half empty country.

Instead of a point, I have questions; and instead of answers, I’d be happy to settle for more, better questions. My instincts tell me that political parties should be ideological rather than regional/social/clientelistic—there’s a slew of reasons for this, but I’ll leave that to future discussion. But would ideological parties actually be better for the poor? Clientelistic parties yield ongoing benefits to the poor, would ideological parties in an ineffective government be able to do much better? How long would it take the benefits of a central government public project, a major hydroelectric plant for example, to outweigh the ongoing benefits dispersed by patrons to poor clients? Do the poor count more in a clientelistic system where they can exchange a vote for a concrete benefit, or in a system where they trade a vote for the promise of a better long-term future? Where does their vote have more causal power?

Assuming we want to, could we use constitutions to shape the evolution of political parties? Would it be through electoral institutions—proportional representation versus majoritarian and all the little ways to manipulate each system (there’s a long discussion to be had on this topic)? How else could institutions shape political parties? What other, non-institutional variables matter when it comes to political party formation?

Is it really possible to have control over where society wages politics? If patronage networks are entrenched, is it possible to shift politics from the networks to political parties? If not, how can we achieve second best outcomes—i.e. how can we make patronage networks better at serving the public good and leading to more efficient long-run outcomes?

Is 8 > 9?

It is a useful fiction that the Supreme Court is somehow beyond politics. It is a fiction frequently strained by the actual state of things. 5-4 decisions have been an increasing share of Supreme Court decisions since the mid-20th century, my “favorite” of which probably being that time 5 conservative justices outvoted 4 liberal ones in Bush v. Gore, effectively electing the president on a 5-4 ideologically-driven decision.

So when Senate Republicans decided to refuse any new appointment for the late Justice Antonin Scalia made by President Obama on the grounds that presidents shouldn’t make Supreme Court appointments in election years to allow voters to have a saw, it’s hard to take them at their word. I think it’s fair to say that there’s been a bit of doom and gloom about the potential future of the Supreme Court being hostage to Congressional partisanship. For now, we’re stuck with a eight-justice court, which means a lot of 4-4 decisions. 4-4 decisions defer to the decision made by the lower court, and avoid setting a precedent. But is that so bad? If the rise of 5-4 decisions indicates an increasing trend of partisanship in the court, then adopting an even number of justices prevents decisions more likely to be found along partisan lines from setting legal precedent. Supreme Court decisions are meant to be based on an understanding of the law, after all, so wouldn’t it be neat if we automatically decreased the importance of decisions decided controversially?

There are a variety of problems with this theory:

  • The fact that there are an even number of justices doesn’t mean they will always be split along neat ideological boundaries (and assumes that justices fit along neat ideological boundaries, it’s hard to argue that Scalia and Roberts were functionally identical).
  • The Supreme Court provides a unifying and clarifying influence on American law. The Supreme Court already decides on which issues are matters for states instead of the federal governments. This system essentially delegates constitutional powers to the Circuit Courts based on whatever happens to be considered controversial among the justices.
  • Attempts to decrease the importance of partisan decisions made by the Supreme Court this way just increases the importance of partisan decisions made by lower courts.
  • Making the size of the court a subject of open discussion leaves it open to partisan grandstanding. Imagine if every time a justice dies or resigns, critics of the president argue that the new size of the court is actually better now.

So we probably should get around to getting a ninth justice some time or another.

A few programming notes I want to make explicit: I have a lot of thoughts that I am pretty sure are wrong, but I can’t really help but exorcise them through writing them down. I am still partially laying groundwork here, so I have several threads running on the blog instead of competing for space in my head.

Also, I am invoking Betteridge’s Law of Headlines on my own title.

On Strategic Correlation (or, the Importance of looking like you’re doing something when you sort of might not be)

Let’s get this thing started.

At Andrew’s suggestion, I made sure to listen to this China Africa Project podcast on One Belt One Road. The brief summary is that China wants to build infrastructure and set up agreements to make trade flow more easily across a Eurasian land route (the belt), and a Indian Ocean maritime route (the road). It’s a pretty natural extension of the String of Pearls idea meant to protect maritime trade between China and the Middle East, it’s another excuse for China to build sandcastles in the South China Sea, and it happens to totally exclude the Americas in this sweet new trade opportunity. Must have been an oversight.

In the course of the podcast, the discussion turned to the the idea that China doesn’t intend on directing the entire OBOR project. What this means in practice is that Egypt and Sudan might decide to build a new rail line between their countries, and this rail line could be branded as part of the Maritime Silk Road’s East Africa leg. This is a pretty sweet deal for the Chinese: they get to claim participation in every infrastructure project along the belt and road. This likely wouldn’t come in the form of claiming credit for inspiring each individual infrastructure project. Instead, we might see reports touting the success of One Belt One Road that includes every expanded harbor and new rail line created, regardless of Chinese influence. Here we get to the title: correlation doesn’t imply causation, and the announcement of One Belt One Road doesn’t necessarily mean that every new infrastructure project  between Beijing and London is inspired by One Belt One road. But any given project could be.

The lesson: If you’re reasonably certain something is going to happen, claim you’re interested in making it happen. You might actually get credit for it.