Aganist the Trump Immigration Order

As you have probably heard already, President Trump signed an executive order 1 restricting immigration, and prohibiting all travelers from a number of countries from entering the United States. This order and a number of Trump’s other decisions on immigration and related issues are absolutely terrible. Despite the fact that I don’t really know people who are directly affected by this, there is a level of sadness and anger at this decision that is difficult to get through on writing. This order will directly harm tens of thousands of people. Trump’s other decisions on immigration and trade will increase this to millions of people who could face severe harm from Trump’s decisions. His embrace of xenophobia, isolationism, and protectionism is by far the most destructive set of policy ideas that Trump has. The human cost of these policies is truly terrible and depressing in every way.

Last year, over 36,000 people were admitted as refugees from the areas covered by the Trump order.

Many of these individuals were fleeing the worse parts of the civil war in Syria (and its spill over into Iraq). Many of them were fleeing brutally oppressive governments. The effort to resettle refugees from temporary camps to permanent settlement should be accelerated, not discontinued.

The conditions in the temporary refugee camps that the refugees live in before resettlement are bleak, 2 and drastically unsafe for long term habitation. The temporary tents and structures in these camps lack heating for the winter and proper cooling, they lack effective sewage disposal infrastructure and lack many other essential infrastructure 3 .  Around 70 percent of the refugees living in camps 4 in neighboring countries in the middle east face extreme poverty and or food shortages.

It is clear that many of those who would of been allowed into America were it not for Trump’s order will die as a result of this decision.

Take the widespread malnutrition and food insecurity in those camps. It should be fairly clear how that can affect mortality, and harm the long run health of those in the camps, but let me spell it out.  The academic literature in medicine is clear 5 that long run exposure to the type of food security that the UN has found in the refugee camps is greatly detrimental to the health of those affected. It increases the mortality of infectious diseases. It has widespread comorbidity with many non-infectious diseases.

Or take the spread of infectious diseases that will happen in the camps.

Or take the lack of availability of basic health services in the camps.

In short if those in the refugee camps are not resettled in a more permanent way, then the results will likely be tragic. It would be fair to estimate that the average person not taken in to America due to the refugee ban has had their lifespan reduced by 10 percent.

The national security and economic case is weak to nonexistent. It is extremely rare for terrorists or criminals to enter the country though the refugee process. Even from the most pessimistic estimates of that number, it is clear that many more people will suffer from not letting the refugees in than from any crimes they do commit. Let’s say, that at the most extreme,  allowing refugees in at the current rate throughout Trump’s term would cause one 9/11 level terrorist attack (something that I am nowhere near willing to concede as a fact, but I will entertain to make a point). Allowing the refugees in, when you look at both the lives saved and the improvements to quality of life would probably do the equivalent to saving 14,000 lives. Allowing the refugees in, even by the most extreme negative possibility would of been on the net positive by a substantial margin. And that doesn’t even count the many other non-refugee migrants and travelers that have been excluded by Trump’s order.

And there are many safeguards against letting dangerous people into the country. Despite Trump’s assertions otherwise, we already have “extreme vetting”. It takes years to go through the refugee process. There are background checks. There are interviews. It is quite unlikely that anyone who is dangerous will get through the process.

With the nationality security defence of this, I since a amoral treatment of the lives of foreigners as close to fully expendable. As far as I can tell Trump doesn’t really consider the costs that his policies will have on anyone but his constituents. He doesn’t care about the foreigners who will die as a result. He doesn’t care about the non-americans who will be impoverished due to his policies. He is willing to throw away an almost unlimited number of lives to meet his domestic policy goals. He doesn’t care about all the people who will be hurt and maimed.  It’s a mindset that I have a very hard time understanding. And it’s hard to believe that it comes from anywhere but racism and callousness and xenophobia.

And then there are the non refugee migrants who will be affected by the Trump order. Students. Parents. Children. People who are just looking for a better life. (CNN did a series of interviews 6 that show some of those stories, and I recommend reading that article)

And all of them will have their world upturned. For nothing but an act of pointless fear that turned into cruelty.

On Compromise and Obstruction

With the Trump presidency there has been a lot of newfound talk about the matter of the role of compromise versus obstructionism. There has been more than a little bit of inconsistency from both sides of aisle on this subject matter.

In this post I will discuss some of the dynamics that I feel exist here.

  1. Compromise should be defined in terms of mutual benefit, not the ideals of one of the parties
  2. Something is only a compromise when both sides benefit as per their own interests and values
  3. A lot of time you get policy proposals that are really false compromises, ie, where a policy proposal is made that is a slight concession relative to the best possible outcome of one side, but which is nowhere near mutually beneficial.
  4. I would define obstruction as one side pushing outcomes that are mutually harmful as an attempt to damage the other side
  5. An additional possibility is an impasse where politicians involved have sufficiently different opinions that a compromise is not anywhere near possible, but nether side can really be blamed for it.
  6. Politicians of both parties like to blame each other for the problems of the country, and like to claim that they are uniquely willing to come to the negotiating table.

Defining Compromise

The only really impartial way to look at what compromise in politics should be is by analyzing the status quo and mutual benefit from a policy. Mutual benefit, relative to the current system is at the core of compromise. This often means that one side may have to give up something, so, that the whole bit of legislation is beneficial for everyone involved. Indeed, the core of compromise is making concessions that turn a policy that may be beneficial to some into a policy that is beneficial to as close to everyone as possible.

The concept from economics of the difference between a Kaldor-Hicks efficient outcome and a Pareto efficient outcome is perhaps somewhat useful here. A Pareto efficient outcome is one that benefits everyone involved. A Kaldor-Hicks outcome is a potential Pareto efficient outcome. Ie. one that could benefit everyone involved — if an ideal redistribution happened after the policy change. Imagine a town with a factory that emits a lot of smoke and pollution. The people of the town would mostly be cheerful to get rid of it. The workers at the factory, however, like their jobs. The cost of the pollution is so high that it exceeds the value of the job loss that would come to the workers. Imagine the case of 1) a lawsuit that got an injunction that forced the factory to shut down without anything being done for the workers at the factory, and 2) a situation where the factory was shut down by a law that also paid for the workers’ job training in a new field of employment and compensated them for the job loss. 2 is a Pareto efficient outcome — everyone gets rid of the pollution but the workers don’t face disproportionate economic harm. 1 is the Kaldor-Hicks efficient outcome — the workers are harmed, but there is enough benefit from the change to go around that the workers could theoretically be compensated (but were not).

Compromise in politics is trying to create more political outcomes that are like option 2, and fewer that are like option 1.

This type of compromise seeking is widely beneficial, helps to reduce hyperpartizanship, and makes the overall political system more functional. An emphasis on compromise helps to defuse some of vitriol of politics, by making sure that there will be relatively few losers from a policy change. It helps to defuse some of the tensions that come from the possibility that people will start to think that the political system is being used as a tribalist weapon by the members of one party versus the members of the other party.

I’m not saying that every policy decision should be shoehorned into an attempt to find a mutually beneficial compromise. There are, of course, a number of political issues that are zero sum in a way that means that one side will by nature be on the losing side. However, it is something that should be done a lot more than has been done in the past, and I also believe that the decline of serious attempts to build compromise are a big part of why political tribalism has ascended. The goal of a lot of politicians is to destroy the other side, not to attempt to build community. Activists and politicians often develop the mindset that the interests and values of the other side have no value, and thus can be disregarded with malice. Therefore, people — often rightly so — see an imminent personal threat in their political opponents. Building compromise is at the core of breaking this pattern.

Fake Compromise

What compromise isn’t is an a concession relative to your own views that still leaves the other side a lot worse off. That’s a fake compromise. And it happens a lot with highly controversial culture war type issues like abortion or gun control. And this can create a lot of nasty issues that potentially break political discourse.


  • The nibbling at the edge issue. If there are two politically powerful groups at the extremes of an issue, you can get a situation where, due to the existence of moderates and institutional constraints, neither side can win outright, but one of the extremes can incrementally advance through small advances due to the ability to carve out groups of moderates and also due to the ability to paint the other side as unwilling to compromise. Eventually, the small victories will add up to something that starts to look at lot like total victory.
  • The shelling point issue. On some issues, there is reason to believe that the government will consistently make sub-optimal issues. This can be dealt with by creating a strong taboo/institutional constraints against legislating on these issues, but, if those constraints are weakened then the floodgates will open.
  • The overton window shifting issue. The proponents of $policy have values that, if applied consistently, would suggest a much broader change in policy. Due to status-quo bias they aren’t yet advocating this, but the range of policy discussion will gradually shift in that direction.

These types of dynamics happen a lot with both abortion and gun control. You have “reasonable” limits pushed as a “compromise”, with the long-run possibility of a radically altered long-run political dynamic. And when politicians try to meet their goals in such a way, but have the gall not to admit it by calling it a compromise, it can cause a lot of damage. I’m not saying that politicians should refuse to entertain the thought of making policy in those areas. And indeed, with a lot of those culture war issues there isn’t really room for a true compromise. But they should realize and admit that a lot of people — for quite possibly good reason are going to not be happy with those policies.


My working definition of political obstructionism is doing actions that make both sides involved worse off (or which make the political system as a whole less functional) in order to force a political concession that makes one side worse off.

I think the debt ceiling controversy was a good example of this. A system where the appropriations for spending were made, but the debt needed to fund that spending were not, the result would of been wildly chaotic. And thus it would be very bad to use that as a negotiating ploy.

Impasse and blame

It’s a feature of the American political system that Congress and the President are elected separately, and thus there is no reason to suspect that legislative and executive branches will be in alignment with each other. This means that it is more than possible that Congress and the President will have wildly different views about what good policy is (as evinced by the events of the last eight years). When they are at loggerheads about good policy then you get an impasse where noting can get done.

The government shutdown was quite different from the debt ceiling crisis in that it was more due to an impasse and difference in policy goals between the branches of government. The President wanted funding for the ACA and related measures, Congress didn’t. And because both Congress and the President are needed to get policy passed, the system was at a gridlock. I don’t think it is very useful to say that either side “caused” this impasse. Because it was the mutual disagreement in policy between both of the participants that caused it. But it does show how both sides will try their hardest to cast blame on each other.

Edits: Fixed a few typeos and removed one sentence that didn’t really fit well with the overall context.

On The Minimum Wage

I’ve been planning to do a fairly lengthy post on my thoughts about the minimum wage for a while. So, here it is.

Q: Do you think a minimum wage is a good idea?

A: Yes. There are a number of good arguments in favor of a minimum wage in some form. I’m particularly fond of economic models that show that employers have some degree of price setting power (and some of the empirical evidence for these models, famously Card and Kruger). My overall opinions on it are thus

  1. It’s better to look at the minimum wage as a backstop to excessive market power in the hands of employers, than to treat it as mostly an anti-poverty measure. A minimum wage increase would be vaguely similar in effect in increased unionisation, or other ways to limit market power of employers.
  2. The overall seems to be little impact on the overall unemployment rate due to current minimum wage.
  3. However, effects on the employment of subgroups of the labor force (young and or, low skill workers) seem quite strong, even if the unemployment rate does not budge overall (due to increases in employment in other sectors). This is to some degree a direct prediction of the price setting power model of minimum wages.
  4. There is some room to increase the minimum wage (~$12/hr) but the $15/hr and $20/hr proposals are excessive.
  5. While a BIG or a similar anti-poverty program may be a good idea, there are strong arguments for having both a BIG and a minimum wage in effect simultaneously.
  6. The dynamic around discussion of the minimum wage is not good. There are actually good arguments for it, but those arguments are complex enough that few people understand them, so, most of the arguments you hear IRL are garbage.

Q: What do you think of the state level $12/hr minimum wage initiatives that passed in the recent election?

I voted yes on the one in my state (I live in Arizona) because it is in the fairly reasonable $12/hr (10 initially increased to 12 over time) range for the min wage,making it within the range of estimates for the optimal minimum wage, and because the paid sick leave policy that was also part of the initiative was very positive.

Part 1: Theoretical arguments

Q: What are the most common models of the impact of the minimum wage?

A: This article does a good job at explaining the textbook models of the minimum wage (supply/demand graphs shown below come from there). Really, I recommend you read that whole link, but I’ll summarize the core points here.

The simplest model imaginable might be a supply and demand equilibrium model — which reflects what can be predicted to happen to the market under ideal competition.

(Image from )

Without a minimum wage the market will reach an equilibrium wage based on the level of demand for the labor of workers at every possible wage, and the willingness of workers to work at every given wage. Now, let’s redo history and imagine that Hillary won the election and was able to put a $12 minimum wage on this market. Notice that the number of workers willing to work in this model at the $12/hr wage now exceed the number of workers that employers are willing to hire. Thus, you can expect an increase in unemployment.

Most arguments against the minimum wage in the economics literature take roughly that form.

Now, lets imagine that really, the equilibrium wage was $10/hr, but employers have a lot of market power here (in the extreme case imagine a small town where a large part of the community works at a single factory, creating only one buyer for a certain type of labor) — and thus wages were depressed. If a $10 minimum wage is enacted you get a different effect.

In this case the number of workers who get hired actually increases. Employment goes up! The employer was increasing profit at the expense of overall output.

This right here is one of the most powerful arguments for the minimum wage.

Q: The whole thing seems arbitrary. Why can’t this be used to justify a $20 wage, a $30 wage, a $1000 wage? Where is the natural limit here?

A: There is actually a very simple way that the price setting power model can be used to predict an optimal minimum wage.

The minimum wage should be set at what the equilibrium wage would have been set at were there not the presence of the price setting power. If the government sets the minimum wage above that level, then there will be an increase in unemployment (because the situation pretty much degrades to the first model in that situation), but otherwise there will not be that disemployment effect.

Q: But labor isn’t an uniform commodity. The equilibrium wage for some workers is going to be lower than the equilibrium wage for other workers. Won’t you get some subgroups of the labor market unemployed?

A: Yes, very possibly. And this is one of the stronger arguments against the minimum wage, despite the possibility of price setting power. You can easily construct a situation where a minimum wage that causes little overall unemployment still unemployed a lot of young or low skill workers. There is a lot of empirical evidence for this . The unemployment of the worse off subgroups, would however, in this case be offset by higher employment in other subgroups of the labor market. And policy makers should (but don’t) take this into account when setting the structure of the minimum wages.

Q: I still don’t get how a minimum wage increase can increase employment in some sectors of the economy. WTF?

A: Here is an analogy.

Consider a power company that is a monopolist. Let’s say that in this market, were it competitive, would come close to the nationwide average power rates of 12 cents per kilowatt-hour. But this isn’t a competitive market. The power company is effectively a monopoly. Its out of the question for someone else to easily set up their own power company. Off grid solar is prohibitively expensive, and thus not a serious competitor.

The power company can charge well above 12 cents per kilowatt-hour. If the power company charges too much demand will sink (ie. due to consumers conserving energy) sufficiently that the company loses money. But it will almost certainly be able to charge a bit extra.

Fleecing incoming! Let’s say that the company finds that its profit optimising price is 20 cents per kilowatt-hour. If a state utility commission puts a price control on the power company (and this type of issue is why utility commissions exist and impose price controls) forcing them to charge no more than 12 cents per kilowatt-hour the impacts will be positive. Prices will fall for consumers, and the amount of electricity sold will actually increase (consumers will back off from the energy conservation efforts forced by the high price for electricity).

Now, if the commission goes too far and, lets say, forces the price down to 9 cents per kilowatt-hour, then the power company will find that it cannot really produce enough electricity at that price to meet consumer demand, and there will be a shortage (more demand than supply).

The labor markets under a price setting power model work in a fundamentally similar way. A powerful employer can optimise its profit by underpaying workers, at the expense of reducing output. A minimum wage that is not excessive reverses this. But an excessive minimum wage will cause unemployment (technically a labor surplus, but equivalent to shortage in the power company example).

Q: The whole mainstream supply and demand model is crap. There isn’t really an “equilibrium price” that exists outside of the action of individuals trading under market conditions. Go read

A: The above linked post from the Mises Institute shows a generic version of an argument that I’ve seen a few times in the minimum wage context (usually from conservatives and libertarians). The author writes “The price is just given. In brief, both consumers and producers react to a given price. But who has given the price? Where has the price come from?

The price is not just given. It derives from the demand/supply curves. The demand/supply in turn come from consumer preferences. The willingness to buy/produce at a given price is where consumers get their say. The shape of the curves is a model that reflects this willingness. The aggregate decisions of all the consumers/producers in the market result in the demand/supply functions and the resulting equilibrium price.

You can think of this in terms of an idealised auction, or a situation where a large number of firms set their prices, and consumers/workers (who in the idealized world have perfect information and no transaction costs) respond.

Q: Why wont, even in the ideal competition case, demand stimulus neutralize the effects that a minimum wage can have on employment?

During a depressed economy the increased consumer demand due to the wage increase can reduce unemployment. The problem here is that this applies only in very particular conditions.

  1. The economy is not at full employment
  2. The central bank is at the zero lower bound (otherwise it could lower interest rates, thus stimulating the economy, and bring willingness to borrow money in balance with the savings rate)

Part 2: Just wage arguments

Q: Why should an ambulance driver, who helps a large number of people, will post-minimum wage earn the same as a grocery bagger. It seems unfair.

There isn’t an extrinsic “just” or intrinsically optimal wage. The optimal wage (in terms of producing the best consequences if the wage is set at that level) for a sector of the economy is set by 1) the number of people who are willing to work in a sector at a given wage, and 2) the demand for workers in that sector at the given wage. The point of variation in a wage rate is to efficiently allocate labor and other scarce resources between sectors.

Q: No one should work and still be poor. We need a living wage!

All throughout the political discourse, you will see the opinion that wages and labor are separate from the pressures that affect all other things in the economy. In a normal market prices are subject to the usual situation of supply and demand. And the prices fluctuate according to how many people want to supply a good, and how many people want to buy the good.

The optimal value for the minimum wage is set, not by the overall costs of living, but instead by the same market factors mentioned above. If the value of the “living wage” is above the optimal value of the minimum wage, then effects on employment will be sufficiently severe as to create more poverty than before. If additional efforts to reduce poverty are needed then they should be done through other programs.

Supply and demand isn’t the only market force that also applies to labor. A business will only hire someone if it sees a profit from that action. But at some point there is only a certain amount of income that a worker can generate for a business. If the wages required to hire a worker exceed the level where it is profitable to hire then the worker won’t be hired. Oh. And there are substitutes to hiring workers. A lot of low-wage workers in retail can be substituted by computer kiosks — if it becomes profitable to make the substitution. Labor markets cannot avoid being exposed to the forces that other markets are, but a lot of arguments for the minimum wage of the living wage form require suspending disbelief on this matter.

Part 3: Empirics

Q: What do economists think about the empirical evidence for the price setting power model of minimum wages?

The economics community is clearly divided on this issue.

A literature review by Hristos Doucouliagos and T. D. Stanley, Publication Selection Bias in Minimum-Wage Research? A Meta-Regression Analysis, found that, after taking publication bias into account, the empirical studies were roughly evenly divided on both sides, such that the adjusted average effect is on the net zero. When the data from the various studies are put together, no net disemployment effects is seen. Although a number of individual studies and authors do find a disemployment effect.

Card and Krueger’s meta-analysis of the employment effects of minimum wages challenged existing theory. Unfortunately, their meta-analysis confused publication selection with the absence of a genuine empirical effect. We apply recently developed meta-analysis methods to 64 US minimum-wage studies and corroborate that Card and Krueger’s findings were nevertheless correct. The minimum-wage effects literature is contaminated by publication selection bias, which we estimate to be slightly larger than the average reported minimum-wage effect. Once this publication selection is corrected, little or no evidence of a negative association between minimum wages and employment remains.

This points towards a largely neutral effect of current minimum wages on employment.

A literature review by Neumark, and Wascher (authors that are skeptical of the minimum wage but well within the mainstream of the economics world), Minimum wages and employment: A review of evidence from the new minimum wage research, is more skeptical of the overall employment effects of minimum wages, and found substantial subgroup effects on unemployment.

We review the burgeoning literature on the employment effects of minimum wages – in the United States and other countries – that was spurred by the new minimum wage research beginning in the early 1990s. Our review indicates that there is a wide range of existing estimates and, accordingly, a lack of consensus about the overall effects on low-wage employment of an increase in the minimum wage. However, the oft-stated assertion that recent research fails to support the traditional view that the minimum wage reduces the employment of low-wage workers is clearly incorrect. A sizable majority of the studies surveyed in this monograph give a relatively consistent (although not always statistically significant) indication of negative employment effects of minimum wages. In addition, among the papers we view as providing the most credible evidence, almost all point to negative employment effects, both for the United States as well as for many other countries. Two other important conclusions emerge from our review. First, we see very few – if any – studies that provide convincing evidence of positive employment effects of minimum wages, especially from those studies that focus on the broader groups (rather than a narrow industry) for which the competitive model predicts disemployment effects. Second, the studies that focus on the least-skilled groups provide relatively overwhelming evidence of stronger disemployment effects for these groups.

And literature reviews involving subgroup effects have otherwise produced strong results.

A literature review by Brown, Gilroy, and Kohen,The effect of the minimum wage on employment and unemployment: a survey, looked at the effect of minimum wage changes on youth unemployment. They give an overview of the theoretical models of the issue and then they look at studies of the empirical evidence. They found that most studies suggest small increases in unemployment in response to an increase in the minimum wage, and substantial numbers of youth will drop out of the labor force altogether. The data is ambiguous on the issue of whether whites or minorities will be affected more.

Q: What do you think this means for policy?

This indicates that the evidence for the ability for the price setting power effect to offset a lot of the negative impact of minimum wages. However, the data available also tends to confirm the possibility that minimum wages will be net negative for substantial subgroups of the labor market. However, there are some methodological limitations on the subgroup studies.

Q: Those studies are about minimum wages increases that are near the current level. What about fight for 15?

Most economists are negative about the $15/hr minimum wage idea, as one survey shows.

Key Findings: Nearly three-quarters of these US-based economists oppose a federal minimum wage of $15.00 per hour. The majority of surveyed economists believe a $15.00 per hour minimum wage will have negative effects on youth employment levels (83%), adult employment levels (52%), and the number of jobs available (76%).When economists were asked what effect a $15.00 per hour minimum wage will have on the skill level of entry-level positions, 8 out of 10 economists (80%) believe employers will hire entry-level positions with greater skills.When economists were asked what effect a $15.00 per hour minimum wage will have on small businesses with fewer than 50 employees, nearly 7 out of 10 economists (67%) believe it would make it harder for them to stay in business.


Compare that to polling on smaller minimum wage increases.


Q: Could there be effects on long term employment growth, instead of a short term disemployment effect?

Yes. Some research has shown this. Although, this is much less well studied.

Q: What is the optimum value for the minimum wage?

A number of studies have estimated the optimum value of the minimum wage to be at around 40% – 50% of the median wage — roughly $12/hr . See, for instance, this paper

Q: Is there a good example of a case where the minimum wage was set excessively high in a way that created a noticeable disemployment effect?

Yes. About a decade ago Congress imposed the national minimum wage on Puerto Rico (which has a much lower percapita GDP than the mainland U.S. This created a minimum wage that was 77% of the median wage — relative to labor market forces the highest minimum wage in the world. See, this paper

This paper seeks to answer these questions using diverse bodies of data on employment and earnings in Puerto Rico and on the employment and earnings of Puerto Rican migrants in the United States. It reports the following findings. (1) The U.S.-level minimum altered the distribution of earnings in Puerto Rico to an extraordinary extent, creating marked spikes that dominate the earnings distribution. (2) Imposing the U.S.-level minimum reduced total island employment by 8-10 percent compared to the level that would have prevailed had the minimum been the same proportion of average wages as in the United States. In addition, it reallocated labor across industries, greatly reducing jobs in low-wage sectors that had to raise minima substantially to reach federal levels.

And this paper

The single most telling statistic in Puerto Rico is that only 40% of the adult population – versus 63% on the US mainland – is employed or looking for work; the rest are economically idle or working in the grey economy. In an economy with an abundance of unskilled labor, the reasons boil down to two. Employers are disinclined to hire workers because (a) the US federal minimum wage is very high relative to the local average (full-time employment at the minimum wage is equivalent to 77% of per capita income, versus 28% on the mainland) and a more binding constraint on employment (28% of hourly workers in Puerto Rico earn $8.50 or less versus only 3% on the mainland); and (b) local regulations pertaining to overtime, paid vacation, and dismissal are costly and more onerous than on the US mainland.
Part 4: Alternate policy options

Q: What other mechanisms than minimum wage do policy makers have to help poor people?

  1. Transfer payments (in the extreme case a Basic Income)
  2. Target employment programs.

We don’t need minimum wages to alleviate poverty, and they have arguably been quite ineffective at it. Even though they are very good at offsetting some of the effects that market power in the hands of employers have created.

A basic income policy that would give all adult citizens an unconditional payment that would be sufficient to take them out of poverty.

Or you could make the MMTers proud and enact a nation wide jobs program — possibly one similar to the Great Depression era Works Progress Administration, but also possibly one created by subsidising private employers — could be created with the task of guaranteeing a job at a reasonable wage to all willing and able to work. This type of jobs program would serve a similar role to a minimum wage, but it would not increase unemployment because the workers that would wind up unemployed under a higher minimum wage would instead wind up on the government payroll. It could be one possible mechanism to alleviate subgroup employment effects.

The combination of the two programs would serve to eliminate poverty, and would give all individuals a chance to earn a good income.

Q: Would a BIG alter the optimum minimum wage, or even make it unneeded?

Probably not. When you look at the minimum wage from the perspective of being a way to offset price setting power held by employers, you can see that the minimum wage really serves a different purpose than a BIG. A BIG is a highly effective way of eliminating poverty. But it doesn’t really do much to take care of labor market power held by employers.

A BIG may make potential employees less willing to work for low wages. But note, this is an effect on the supply of labor. Not an attenuation of the power of employers in a way that is germane to setting the minimum wage. Go back to the supply and demand curves that I inserted in part 1. If the supply curve shifts, the equilibrium wage will go up, along with the actual wages that employers pay (even under the imperfect competition model).

But the employer with price setting power still has the ability to force wages below the market equilibrium wage. And indeed, if the existence of a BIG increases the equilibrium wage for low wage labor, then the optimal minimum wage may actually increase.

Thoughts on political polarization

In this post I will sketch out some of my thoughts on two questions.

  • What has caused the recent uptick in political polarization?
  • Why has the right polarized more quickly than the left?

One fact that has been made clear by the recent presidential election and its aftermath is that the country is very divided, and polarized, with extremism taking center stage in American politics. But it is also clear that, while a progression towards extremism has happened to some degree on the left, it has unfolded much more quickly on the right. The most extreme parts of the GOP as represented by Donald Trump have been able to, for all intensive purposes, take over the party. In the Democratic Party, however, the establishment wing of the party has been able to hold its ground for the time being.

Why is the left less susceptible to the influence of political extremism?

It’s not so much because of the lack of extreme voices, as those exist in legion, as discussed by this recent article. It’s more that for various reasons, the institutions of the political left have been much better at keeping its worst parts out of the mainstream.

Here is my model of this. I think a lot of the increase of extremism on all sides of American politics is due to the increasing ability for people to isolate themselves from serious engagement with other viewpoints. The Internet allows you to bubble yourself in media sources (often severely lacking in rigor) that reinforce your viewpoints. Cable television is also to do the same thing to a lesser degree. New media, unlike old media is therefore loaded with a tendency to be inherently ideological. In the old media landscape this was limited to some degree by the mere scale and cost of setting up a media outlet. And then there is what author Bill Bishop called the big sort, people are moving to communities and entering social groups where people are largely like minded on political issues, further eliminating their ability to engage with people who are not like them or do not share their views.

This has a number of nasty effects. The first is that it eliminates a lot of potential room for productive discourse. If you don’t know any people who have a viewpoint it is hard to understand it. It is unlikely that you will interact with arguments against it. You will be trapped in a bubble. And this leads directly to the second major issue, what Nat Greene and Erik Fogg call the wedging effect. It becomes a viable strategy for politicians and media outlets to try to vilify their political opponents and to use some of the most incendiary political issues in the political system as a lever towards this.

You can portray the other side as a imminent threat to your most core values, and thus fuel the flames of political tensions. The fact that people are less likely than ever to actually interact with and understand viewpoints other than their own make this much easier. It’s easier to think that an abstraction is evil than an actual person that you know in real life. It’s harder to see though strawmanning of ideas if you have never seen the real idea explained to you before. And this cycle rapidly becomes self reinforcing, in a dangerous runaway feedback loop.

Greene and Fogg also discuss another core part of the veer towards extremism. People with extreme viewpoints are much more likely to participate in politics (as shown in the two graphs below). This has two effects. One, it makes politicians and media outlets even more willing to cater to them. Two, it biases political decisions towards their worldview.

Here is where I form my hypothesis on why Democratic Party political institutions have been more resilient towards being taken over by extremists. Note that, while the effect shown in the second graph is still there for both liberals and conservatives, the extremism/political participation effect is noticeably weaker for liberals than for conservatives. This means that, even though extremism is growing, the new extremists, when liberal will not be as likely to try intensely to get their views put into policy. The most extreme Democratic voters, therefore cannot as easily take over their party’s political institutions as their GOP equivalents.

I can speculate a bit on some of the reasons for this. I think it is likely that a lot of the same factors that can make someone have extreme left wing viewpoints can also make one disengage from the political system all together.

Presidential Election Post-Mortem

Unfortunately, Donald Trump won the election and it looks like he will become the next President of the United States. In this post I will elaborate on some of my thoughts about this.

What Will Happen to Government Policy?

I’m very worried about what Trump will do to international relations. The president has a lot of unilateral authority in that region. At least with domestic policy, there are a lot of breaks on presidential authority. There is Congress (which, even though GOP controlled will be able to filter out the worst Trump ideas). There is the Supreme Court. There are independent executive regulatory agencies. There are state governments. In short to get anything done in the domestic policy sphere you need to cut a deal with others, and get some degree of broad based support. The rich guy who has never been told no probably won’t be a happy camper when he finds this out. And if anything there will be a silver lining here in terms of a renewed push from the left towards a minimalistic interpretation of executive branch power.

International relations are different. The president has enough control over the military to for all intensive purposes start a war on his own (even if Congress is needed to formally declare it). And I don’t trust him to use this power in a restrained and well thought out way. To me Trump seems very rash and impulsive, and is very bad at handling offence. Not traits you want in the chief diplomat of the country. And he has a very militaristic style, which runs the severe risk of getting America into pointless wars.

And Trump can also do a lot of unilateral damage on trade and immigration. He could increase substantially the number of deportations that happen, he could cancel the Obama administration executive actions on immigration, he could pull the U.S out of the WTO and NAFTA. And Congress would have very little chance to stop him, even if it wanted to. And this is very dangerous as immigration and trade are some of the best ways we have to help the global poor. The world needs to become more economically and culturally integrated, but Trump wants to move the world away from global unity and towards nationalism and xenophobia.

But Trump’s ideas on domestic are also very bad, only buffered by their incoherence on many points. He will run the long-run deficit to record levels. He wants to repeal the ACA, which has expanded health care availability to millions, while his ideas for a “replacement” are insufficient at best and incoherent at worst. His presence in the White House will put the brakes on climate change action long enough to make meeting the Paris targets effectively impossible. His choice for Vice President, Mike Pence, has a terrible record on LGBT rights.

Trump will probably get to pick at least two SCOTUS justices in his term. Scalia’s post remains unfilled, and I suspect that there is a very good chance that at least one other justice will retire or die (Ginsburg is probably my guess for the next vacancy). Personally, I would like more centrist judges of the Kennedy or Roberts mold. However, I am worried that he will pick a very partizan court, and this will endanger both the quality of its rulings and its trust and credibility with the public at large, and may have very widespread implications for civil liberties, civil rights, and other important issues.

What does it mean for predicting future elections?

Nate Silver’s projections for the election held up fairly well, as he was one of the few major predictors of elections to predict a close election. So, this is a major feather in his cap on this issue.And a major egg in the face of the people who predicted a blowout in Clinton’s favor.

There has been a bit of a tension in the political science world between models that have partizan affiliation as definitive in elections, and models that show the candidate as definitive. This election points in the favor of the party-centric models. Trump on many grounds was about the worse candidate imaginable. But he did well. He kept the usual GOP base. This shows that voters are mostly deciding their vote on the basis of party affiliation, and party loyalty.

And there has been a strong party that cried wolf dynamic here. Hyperbolic and extreme attacks on political opponents, have loosened the credibility of those attacks even when they really do matter. Voters respond to what high placed officials and prominent leaders say. And those leaders have in the past damaged their credibility. Also, The wolf cry dynamic has, by conflating legitimate policy disagreement with being evil, damaged the weight people give to actual severe defects in policy and personality.

Racism and Bigotry

There is, of course, a large role for racism and xenophobia in this election. Racism that is tied to protectionism and ultra nationalism has been for a long time one of the few socially acceptable forms of racism to openly express. And Trump’s campaign, as Mitt Romney has put it, has created a dose of “trickle down racism”. Trump has made bigoted views even more within the mainstream.

Trump’s racism and pandering to racists, has made it clear that racism still has a very nasty role in politics. Trump’s hot mic video and the response to it show how the worse forms of sexism still have a strong presence.

This must be acknowledged. And it must be fought very aggressively.

But the left has really screwed up with handling this. A large part of the left was under the impression that it could deal with racism by asserting a strong overton window, and punishing any malefactor that dared to cross it. This 1) caused blowback (cf. political correctness) and 2) didn’t actually fix the issue. Voters minds need to be changed or their (even if extremely wrong and dangerous) will leak into politics. You need discourse and persuasion, or things will go very bad. You can’t just turtle in the corner.


A lot of commentators have noted the strong working class support that Trump got.

On one hand there is a strong sense in which this is both somewhat true and way overstated. The Democrats did do poorly relative to other elections in low income groups. However, there are some serious caveats here.

Trump’s level of support was fairly uniform across income groups, and he did the worst in the two lowest income groups. And in the primary Trump’s voters had a median income of over $70,000, which is well above the average. It was more that Trump inherited the working class parts of the GOP base, he didn’t bring those people into the GOP.

Populism really is the home turf of the white middle class, it was the core of both Trump’s support in the GOP primary, and Bernie’s support in the Democratic primary. And this election was in large part upper middle class white people throwing their tendies over the fact their demographic’s chokehold on the political system is at risk.

Although there is (as evinced by the Michigan results) a strong subset of the working class (particularly white and rural) that the Dems aren’t getting to. There has to be a lot of revaluation of how the Democratic party interacts with that demographic and a change in strategy. And perhaps a change in policy.

But a large part of the issue here is that middle class still has its chokehold and the Dems can’t afford to really make the valuable white middle class demographic worse off, even to help poor people, so tensions between demographics within poor people get stoked, and large scale redistribution towards people facing financial stress (particularly workers displaced by market changes due to automation and trade) is off the table.

And neither the populist xenophobic Bernie wing of Democratic party, nor the in the beltway Clinton wing has any good ideas here. Bernie’s economic policies were mostly about giving free stuff to middle class voters (often with dubious economic justification, and often at the expense of the poor). While the rest of the Democratic party hasn’t even seriously tried to grapple this issue.

And then there is the point that Bryan Caplan made back in the aftermath of the 2012 election “People vote for whoever respects them more”. Where “respect” is less about the effects of substantive policies on the groups involved, and more about the soft interactions involved. It’s about not being blatantly insulting to the groups involved. It’s about giving the appearance of caring. And the left has been quite bad about showing this type of respect, and really doesn’t appreciate the extent to which class can be a very strong axis of oppression on par with race or such. Class is at best secondary issue for the left. And the left has a tendency to be very insulting towards the working class demographic, even when the insults are tangentially at best related to the serious issues at stake (I’m seeing a lot of Facebook posts about “rednecks”) . The right has been quite good about this, and does the minimum amount of lip-service needed to show respect towards this demographic, even if the GOP’s policies are lacking.

And considering what I am seeing on social media, this may not change. I’m worried that the left will get the inverse message from this election and double down the behaviour that resulted in this, such as the hostility towards rural working class people and tendency to ignore class in forming policies. And therefore add gas to this dumpster fire. I had an anti-Trump Facebook friend seriously discuss “political literacy tests” to filter out stupid voters, and I saw a few ironic suggestions of that nature. There is a torrent of posts about “rednecks”, and of course, the condescending low-info voters voting against their self interest type posts. And much less pointed at trying to understand the issues that drive them to vote for Trump-type candidates.


See my response to an ask on tumblr I received in response to this post

The ask

“Class is at best secondary issue for the left.” look, you don’t understand us. This is exactly what some of us have been fighting against. We were screaming doom about this narrative. We predicted right-wing populism taking up our turf. We weren’t heard.

My response

I want to clarify a bit here. I wasn’t really referring to your part of the left, of course class is your bread and butter. (I think you are referring to this post of mine on my mainblog )

I was referring to the mainstream of the Democratic Party, and the SJ/campus activist part of the left.

And I will readily concede that it is very troublesome that the mainstream/clinton part of the Democratic Party responded to the defeats of the 70s and 80s by largely abdicating economic policy and stopping any serious attempts at helping the poor.

There are many people who are suffering severely due to the effects of automation and globalization and even though both are beneficial in the broad scheme of things, little has been done to help those that are suffering and make those changes as universally beneficial as possible. And even those changes aside, there are lots of poor people who need assistance. And who aren’t getting it. Despite my support for Clinton, I have very major issues with her policies (or lack of policies) on those matters.

And the SJ element of the left has refused to acknowledge and internalize how strongly class can be an axis of oppression (perhaps because a lot of the academics and writers involved are quite privileged in that regard). It has a nasty tendency to try to fix oppression by beating up on the most disadvantaged of the privileged group (instead of fighting those with actual power or unearned privilege) , as can be seen in its interaction with rural working class whites, or with neurodiverse individuals, or in many other cases.

And this means that the right has won on economic issues by default with little contest from the left. And its pernicious attempts to roll back the economic safety net have taken the day. I’m not a libertarian or a conservative, and I do think that the victory of the right is very troublesome and will hurt a lot of people.

But here is where I think we disagree. I don’t really think that rest of the left has offered a good alternative.

The Bernie segment of the Democratic Party is still a different type of isolationist populist. Bernie’s economic policies are centered on helping the upper middle class, not the working class or the poor. Relative to the ACA the poor would net lose under Bernie’s health care proposals. Bernie’s plan included a large flat payroll tax increase at the federal level, and put expenses on the states that would be covered by sales taxes, and would give poor very little that the ACA hasn’t. His education plan feeds the bubble elements of the higher education system, and throws money at many people who are already quite well off.

And the more traditionally socialist harder core elements of the left, I believe are making roughly the inverse mistake of the mainstream Democrats. The mistakes the left made during the 70s did serve to show that on grounds of efficiency markets fundamentally work, and thus to some degree they have to be kept at the core of the economy. Fixing the nasty distributional impacts of markets, and fixing market failures are extremely important goals (again I’m not a libertarian). But markets should exist in some form.

On immigration and social stability

This is kind of a response to this part of the recent never-Trump essay by Scott Alexander, plus similar ideas I have heard elsewhere 1.

The one place where Clinton is higher-variance than Trump is immigration. Clinton does not explicitly support open borders, but given her election on a pro-immigration platform and the massive anti-Trump immigration backlash that seems to be materializing, it’s easy to see her moving in that direction. If you believe that immigrants can import the less-effective institutions of their home countries, …. or cause ethnic fractionalization that replaces sustainable democratic politics with ethnic coalition-building (unlike the totally-not-ethnic-coalition-based politics of today, apparently?), that could potentially make the world less functional and prevent useful technologies from being deployed.

In a later post I will deal with the assertion that Clinton may substantially expand immigration in a way that is radically different than the status quo (I think that she seems to pretty much want immigration policy kept as-is with perhaps more visas given to refugees).

In this post I will discuss why immigration at anything close to current levels for the most part has not seriously damaged the functioning of society, and if major policy mistakes are not made immigration will not be likely to pose a substantial risk to social stability in the United States.

The caveat is that this is only true as long as a reasonable effort is made to avoid a marginalized underclass, like has plagued immigration in Europe. Fortunately, avoiding the underclass risk has been the one thing that the United States has been reasonably good at avoiding with immigration in the past. Issues that have been often taken as severe negatives of immigration, like multiculturalism are actually relatively neutral to positive for the U.S and are necessary for integrating the immigrant populations into the broader U.S society. And if anything Trump’s proposals are much more likely to bring the country closer to the dangerous marginalized underclass situation.

A 2010 paper by Kesler et. al, Does Immigration Erode Social Capital? The Conditional Effects of Immigration-Generated Diversity on Trust, Membership, and Participation across 19 Countries, 1981–2000 published in the Canadian Journal of Political Science 2, looked into the effect that immigration has had on social stability as measured by a number of factors. They found that whether immigration damages social stability of the host country is very conditional on how the host country handles other policy issues relating to immigration. In some cases immigration can cause damage to social stability, but in other cases it can improve it. Multiculturalism and pluralism towards ethnic minorities actually is strongly determinative in preserving social stability in the face of immigration. A strong economic safety net is also very predictive of the effect that immigrants will have on society with a stronger safety net correlated with more social stability.

It is worth noting here that, despite the stereotype, the population of the United States is actually, by opinion polls and other metrics, more tolerant and more pluralist than most of Europe. This is discussed in a recent article 3 in The Economist.

This means that the U.S is in the zone where immigrants will tend to be integrated into the broader society instead of excluded, and thus immigration will be a boon for society and social stability.

Herreros et. al. in a paper titled Social Trust, Social Capital and Perceptions of Immigration published in the journal Political Studies 4 , contends that a lot of the perceived negative effects that immigration can have on social capital in many cases are due to exogenousity issues. Namely, immigrants tend to 1) have low social capital, and 2) migrate to countries where social trust and social capital are high. The effects on the social trust and social capital of natives in the host countries is very low, and the social trust and social capital of the immigrants could actually increase.

The evidence that multiculturalism is inherently negative is weak.

Let’s look at some of the ways that immigrants can change the surrounding culture, such as bilingualism, on all of natives, immigrants, and society. Immigration has the potential to create an economic advantage to becoming bilingual — ie. the immigration creates a Spanish speaking consumer base, and businesses that can sell to it can profit from its existence. However, there is no reason to suspect that this is a zero sum result that comes at the cost of the typical non-immigrant. The Spanish language market would not exist were it not for immigration. It therefore creates new value that can be obtained by the bilingual, but does not net destroy value. Additionally, the economic value of being a native speaker of English also increases, due to that skill becoming less common proportionally in the population, and due to possible economic uses of that skill.

A 1997 paper The economics of multilingualism: Overview and analytical framework published in the Annual review of applied linguistics 5 suggests that bilingualism is likely to have positive net economic effects over monolingualism. However, the authors report that data on the economy-wide effects of multilingualism are less well studied than individual scale effects on income.For countries that have a history of multilingualism, the administrative and educational costs produced as a result multilingualism are material but not prohibitive. This was estimated at approximately 0.5 percent of GDP for Switzerland (where four languages are commonly spoken). Data of this kind for the United States tends to be scarce.

Bilingualism produces a large benefit for personal earnings for both immigrants and non-immigrants across multiple countries, including the United States according to the paper The endogeneity between language and earnings: International analyses published in the Journal of Labor Economics 6. The effect is stronger for immigrants who acquire the language dominant in the country they migrate to, than it is for non-immigrants who acquire a second language.

Next let’s look at political institutions. One of the models that I think can best predict the impacts that immigrants will have on political institutions is the selectorate model of politics, developed by Mesquita et. al. in their academic book The Logic of Political Survival and in their popular book The Dictator’s Handbook. The Logic of Political Survival contains the discussion of immigration, but The Dictator’s Handbook is very recommended. In this model, there is a tension between a politician’s choice to provide broad-based services that benefit most of the population, and private goods that benefit the supporters of the politician. The same money can’t be used for both. Rulers in both dictatorships and democracies need to get support from others. The main difference in political systems is the number of people in the winning coalition of supporters that the politician needs to remain in power 7 . The cost of providing private goods scales with the number of people in the winning coalition and the cost of providing public goods scales with the size of the total population of the country. This means large coalition governments (ie. well-functioning democracies) are much more likely to spend on public goods than small coalition governments (ie. dictatorships).

Things become interesting when you add immigrants to this model. If you have a large number of immigrants that are excluded from effective political participation, the winning coalition size as a percent of the population declines, and thus politicians become more likely to allocate resources to economically inefficient private goods and the overall functioning of political institutions is damaged. However, if it is relatively easy for immigrants to gain voting rights, the winning coalition size (in raw numbers) increases, and thus the government can be predicted to become better functioning.

Again this is a repeat of the general rule of avoiding excluded underclasses of immigrants. In my opinion, this is one place where the U.S has failed with new immigrants, but the existence of jus soli citizenship in America has reduced this issue with the children of immigrants