Monday assorted links

1. Don Boudreaux responds to Dani Rodrik on free trade.  And Arnold Kling responds.  And Dan Klein reviews Charles C.W. Cooke’s The Conservatarian Manifesto.

2. A quick way to access what betting markets are telling us about the candidates.

3. The NBA is moving toward wearable tech — are you next?

4. More data on Chinese services, including telecom.

5. Should we adjust for this data, or learn from it?  The temporal evolution of citation counts, across fields.

6. Generalized results about interest rates and banks profits, check out the Chicago Fed link here, Genay and Podjasek, it is #2 in the order which comes up for me, or try this link.  Overall banks are more profitable when interest rates are low, contra some recent blog posts by Krugman.   This effect is especially strong for small banks, which have a good deal of political power.  In any case “Interest rate changes generally have small effects on bank profits…” and that effect is dominated by the effect from macroeconomic conditions, including employment and also home prices.  Here are useful comments from Brad DeLong.


What? Nothing about the Not Very Serious People being re-elected?

Funny you bring it up, The Not Very Serious People got rid of The Not Very Serious Person, and since have become The Very Serious People.

Does this mean you didn't read is Not Very Serious Book?

I like the new color scheme. However that is not fair. The Not Very Serious People got rid of *one* of the Not Very Serious People and now they have become the Mildly Serious People. Partly because they dumped one of their clowns, but mainly because of experience. Slowly the situation they are in is sinking in and no matter now many times they do the sums, the figures don't change.

They are still a clown show. We have not seen the last of their attempts to pull a full sized trombone from their pants yet. Someone is going to get covered in white paint and lemon pie or they will die trying. I expect it will still be the European tax payer.

They didn't dump just one not very serious person. In addition to him, 25 of their most radical MPs left the party and founded the Popular Front in opposition to accepting the latest bailout agreement. Their new party was absolutely crushed in the elections, falling below the 3% line and not managing to elect a single candidate. As a result, Syriza is now much more manageable than it was before, having lost it's most troublesome 1/6th.

But the Greek voters did not vote for any of the responsible parties. The traditional Right and the Socialists got some votes but they are not in power.

So the Greek voters have told Syriza they are fully on board their efforts to avoid paying while ensuring the flow of Germany money. They have just said they shouldn't annoy the Europeans so much - after all, then the Germans might stop paying.

Bottom line is still that the Greek voters think this clown show can actually pull a trombone from their pants and so they won't have to pay.

Re bank profits: generalization from long-run averages is not appropriate in this case. Deposit rates are usually a percent or so below wholesale short-term interest rates. That's not true today, because of the zero bound. With short rates at a handful of basis points, the spread between wholesale interest rates and deposit rates doesn't cover the cost of mailing out a monthly statement (unless the account is very large). As rates rise, banks won't raise deposit rates for a year or so; they are flush with deposits relative to loans, so won't compete with deposits. I have a whole article on the subject on my website.

And Dan Klein reviews Charles C.W. Cooke’s The Conservatarian Manifesto.

A member of the British chatterati, graced with an impossibly impractical education, shows up on our shores and takes a job in the philanthropic sector from which he attempts to sell sh!t finger sandwiches. Libertarian professoriate applauds.

I rather like some of the stuff he's written for NR.

I'd like Klein to talk his friends, when they see a proposal they dislike -- ACA or a higher minimum wage, propose a better alternative showing that they don't object to more low income people having health insurance or low wage workers having higher incomes. I think we could find many many ways to raise aggregate income if Republicans were willing to risk explicitly higher taxes on high consumption people.

1. The past 35 years have been a perfect storm of events that severely test the theoretical benefits of free trade (globalization). First, on globalization, it provided a huge shift of manufacturing from the U.S. (developed countries) to developing countries, creating enormous wealth in developing countries, increasing profits for U.S. companies which sell those manufactured goods (as the result of lower costs), and resulting in lower prices for consumers of those goods in the U.S. Sure, some workers in the U.S. lost their jobs, but consumers in the U.S. benefited greatly (from lower prices). As for the enormous wealth created in developing countries, those at the bottom clearly benefited (even though manufacturing wages are low, compared to earnings from (for example) agriculture, manufacturing wages raised the standard of living above mere subsistence). However, most of the wealth created in the developing countries went to a very few - inequality within developing countries is extremely high. Back in the U.S., as trade barriers were coming down (and as manufacturing jobs in the U.S. were being shifted overseas), the U.S. experienced an enormous shift in public attitudes and public policy, including much lower tax rates for companies and the affluent (though payroll taxes for working Americans increased significantly) and, more importantly, public attitudes against both taxes and government. Such public attitudes are reflected in (for example) the largest U.S. company, Apple, paying very little in U.S. tax by deflecting most of its income to a tax haven by employing aggressive tax avoidance techniques with little or no substance (some call it tax evasion). Not surprisingly, inequality in the U.S. has risen significantly. According to a recently published book, close to $8 trillion is hidden in tax haven countries (Ireland, Switzerland, etc.). Rising inequality in the U.S. combined with rising inequality in developing countries has strained global markets, as the world has experienced a savings glut that has contributed to slow or no economic growth and financial and economic instability. Indeed, some economists believe the world is in for a long period of secular stagnation. In many of the developing countries that have benefited from the shift in manufacturing to developing countries, there is concern that secular stagnation will strain social and political relations, leading to political instability (and a move away from democratization). Rodrik often this concern. So did free trade (globalization) make the world a better place? It's debatable. What can be done? Three choices: First, abandon free trade (globalization) - an unlikely choice. Second, better manage the transition to a truly global economy by adopting policies that mitigate the negative consequences - this would require a shift in public attitudes that is unlikely and result in policies (e.g., higher taxes on tax evaders and the affluent) that the wealthy are unlikely to support (or allow politicians to support). Third, let markets be markets, no matter how painful it may be and for however long it takes - that's the free marketers' approach. Of course, it helps if you have tenure.

Why not ask the moderator to delete this and you re-enter it with paragraphs?

rayward September 21, 2015 at 1:46 pm

However, most of the wealth created in the developing countries went to a very few – inequality within developing countries is extremely high.

Do you really believe this nonsense? I agree that in a country like China inequality is very high. That we all know. But over the last seven years, wages in China have roughly doubled. They do that about every seven years. It is now about $8,000 a year.

The richest man in China is Jack Ma thanks to his IPO. He is worth about $24 billion. So is the next richest guy Wang Jianlin who founded a company called Dalian Wanda. Notice what these two do - they are not manufacturers. Wang is the Sam Walton of China - he owns shopping centers. Jack Ma owns a company that sells stuff over the internet. That is, they are rich because China's consumers are rich. Who the third guy is, I don't know. But I doubt he makes anything (except profits of course).

Jack Ma has about $2 for every Chinese person. Thanks to naive American investors. They *make* (that is, a measure of their income, not their assets which are naturally higher) 4000 times that on average.

According to a recently published book, close to $8 trillion is hidden in tax haven countries (Ireland, Switzerland, etc.).

According to many recently published books, inter-galatic, shape-shifting lizards, who happen to take the form of Jewish people, secretly run the world. Don't believe everything you read.

"inter-galatic, shape-shifting lizards, who happen to take the form of Jewish people,"

Don't be ridiculous, there's absolutely no evidence that our Reptilian Overlords come from another galaxy.

That is what they want you to think

Google "Gini" and "China" and you will find that the gains of growth have disproportionately benefited the wealthy.

The fact that some people believe in a Jewish banking conspiracy is not a counterargument against the $8 trillion figure he cites.

Nathan W September 22, 2015 at 12:18 am

Google “Gini” and “China” and you will find that the gains of growth have disproportionately benefited the wealthy.

I googled Gini and China. I discovered the top 10% now have twenty four times the income of the bottom 10%. Which is pretty unequal. But it does not prove that the gains of growth have mostly gone to the rich. Nor does it take into account someone going from $1 a day to $2 a day has improved his life a lot more than Jack Ma going from $12 billion to $24 billion.

The fact that some people believe in a Jewish banking conspiracy is not a counterargument against the $8 trillion figure he cites.

It is not meant to be. It is a reminder that half-remembered claims from some fringe publication is not evidence. And strong claims require strong evidence.


You are at least 80% (maybe 90%) right. However, statements like "However, most of the wealth created in the developing countries went to a very few" are not factually correct. China is a rather good case in point. Has inequality increased in China since China abandoned real socialism? Clearly the answer is yes. The Wikipedia page on the subject ("Income inequality in China") states, "A study published in the PNAS estimated that China’s Gini coefficient increased from 0.30 to 0.55 from 1980 to 2012". However, the IMF data shows that real, constant currency, per-capita GDP was 15.25 times higher in 2012 versus 1980. By 2015, real per-capita GDP in China was 18.54 times higher than it was back in 1980.

It should be obvious that the gains from an 18.54 increase in per-capita GDP utterly dwarf the losses from an increase in inequality. Is China an extreme case? Yes, China's growth record is essentially the best on earth. However, China has also "enjoyed" (so to speak) above average growth in inequality.

A statistical way of looking at this might be... Can we find any developing country where a majority (50+%) of the GDP gains since 1980 have accrued to the top 1%? I tend to doubt it, but I could be wrong. Conversely, that's probably a true statement for the U.S. since 2000 and quite possibly for longer periods as well.

3. To answer the question, hell no. I can't think of a single activity I do--including weightlifting and swimming to stay in shape--where I'd want to record and sift through data recorded from a sensor. Also, knowing that I'm one of those people who wears a sensor would weigh heavily on my soul.

Anyway, the implications of sensors on pro sports are interesting. Is feedback enhancement a form of performance enhancement? The issues and discussion around performance enhancement are very interesting to me.

Continuously monitoring blood sugar would be extremely helpful for planning diet for exercise. Such monitors already exist.

Why do I need to do that? I eat healthy, I exercise. I've been doing this for years and getting results I like.

One thing I really don't understand about my generation is this rush to overcomplicate everything.

If you are weightlifting and swimming to stay in shape, and not just for its own sake, then you might want to know how to stay in better shape or how to stay in the same shape at lower cost. Or you might want to know how to improve your feeling of well-being during those activities.

Well I guess I'm also doing those things because I enjoy them to some extent. I don't know the ratio between enjoyment/staying in shape.

But like I said, I'm getting results I like. I don't have a need to stay in BETTER shape, at least not through more data about my swimming/weightlifting. (Instead, I could probably stand to stretch more often, eat less when I'm inactive, etc.)

I have a pretty strong ethos about asking "Do I really need this one more thing in my life?"

What if you didn't have to sift the data? The fitness tracking software would just develop a few consolidated measurements from its sensors and give you some bar charts or scores for what you care about, like cardiovascular health, risk of type 2 diabetes, etc.

Current wearables have sensors that produce a lot of junk data, but there are sensors used in research (pulse wave velocity, pulse contour analysis, skin blood flow, endothelial function) that can produce data with good predictive value. I'm working on making one of these sensors into a form that can be low cost and wearable. If this project is totally successful, your wristwatch will be able to indicate when you do unhealthful behaviors like smoking a cigarette, drinking a sugary soda, or eating a high-fat meal. Within minutes, it'll indicate you've done something wrong -- it won't know what you've done wrong, but it will indicate you've done something unhealthful.


If you've done something unhealthful, wouldn't you realise it in the 1st place?


I suppose NZ would, but many other people are in denial. If it's deep-fried in a "healthy fat" like olive oil, some people willingly self-delude into thinking it's good for them. But the objective health sensor will not be fooled. You ignore it at your peril.

Please keep us & Nicholas Carr apprised of future developments. ;)

My great grandfather lived to 104. My grandfather is 91 and still living independently in his house with my 92 year-old grandmother. They eat basically whatever they want, though my grandfather had to cut back his salt about 10 years ago. None of them ever needed a computer to tell them what to eat or how much to move. Millions of other people can probably report similar things about their elderly relatives.

If I had a strong urge to throw away my money so that some snotty hipster in San Francisco could afford his $500/sq.ft apartment or save up for his Tesla, I'd buy a sensor and use it to tell me what to eat and how to be healthy.

If the most relevant parts of it were displayed for me in a really cool and easy to understand infographic, I'd still hate it. In fact I might hate it more.

You are an infopath.

No, I just value equilibrium.

I agree. It's not wholly rational, but the thought of being minutely quantified makes me depressed.

I encourage you to explore that.

#1. The only valid objection I can think of to free trade is if the foreign country allowed the use of slave labor to produce products for export. I feel this way about prison labor as well. Prison laborers are not compensated at market wages and cannot bargain or seek other employment. The contracts for prison labor strike me as highly cronyist too. Basically anywhere that the labor force is under physical constraint is going to have a competitive advantage over other domestic businesses that can't be legally duplicated.

Trading with a slave state is still a utilitarian benefit. If you don't trade the slaves will continue to be slaves, but their marginal product will fall. They may not capture that completely like waged workers, but they'll still internalize some of it. Worse conditions, less food, less healthcare, higher death rates, etc.

The problem is that allowing slavery to be profitable incentivizes slavery. If the slave states have a competitive advantage, they will put non-slave-based businesses out of work, and thus incentivize others to practice slavery. In a highly competitive market, the end result would be that only slavery-practicing businesses would survive.

Also, refusing to trade with slave states could potentially make slavery sufficiently unprofitable to convince the slave states to abandon slavery. Yeah, the slaves would be worse off temporarily, but they might be better off in the long run if the end result is that they aren't slaves anymore.

This is taking the free trade argument to almost Swiftian extremes. It wasn't slavery, of course, but President de Klerk of South Africa himself admitted that sanctions against South Africa were one of the reasons apartheid was dismantled when it was.

You folks are being naive. Any country inclined to use slave labor for serious production, won't care much (either way) about trade restrictions on the use of slave labor. The domestic market will always be large enough to handle the output from slaves.

Any opinion on prison labor?
It seems sort of unfair if one company gets an exclusive contract to hire prison laborers. Everyone else has to compete with them, and there's likely a lot of corruption involved in the bidding and awarding. And the "employees" can't exactly go work for someone else.

This argument is almost certainly wrong. It is tough to imagine the sugar plantations of the Caribbean in the 18th century being profitable, for instance, if it weren't for international trade. The number of slaves in a society is not independent of economic conditions -- if there is a strong global demand for a certain commodity and it can be manufactured or mined using slave labor, the pressure to enslave more people increases.

Re bank profits, levels v. changes:

The question is whether bank's like, profit from, a period in which interest rates are increasing (distinct from profiting from higher interest rates). The underlying claim is that bank assets generally reprice more quickly than bank liabilities.

Or probably really banks just want the Fed to generate higher expected rates while actual rates remain low.

It may be true that on average, low rates can be fine for banks, but the zero bound is a problem because the zero cost demand deposits and low cost MMAs you've invested so much in gathering cease to save you on interest expense, which is part of the reason NIMs and ROAs are at all-time lows.

DeLong's idea of raising the inflation target to 4% is no good for banks either because most banks have mismatched duration - rising inflation gashes the value of of your loans, particularly fixed rate mortgages, which got so many thrifts in trouble in the 1980s.

Most banks make more money when interest rates and/or inflation are declining, but at the same time, all things being equal, they would rather have high rates than low rates because they lend more than they borrow, and certainly they would like to be above the zero bound. (ROAs were much higher in 2000 than they are now in 2015, for example). If you take the long term average, banks are probably more profitable in years where interest rates fall, because of the duration mismatch, but in the longer run, lower rates are not good because they are net lenders and benefit from their base of low cost and zero cost deposits.

It's my understanding mismatch cuts the other way (i.e., that bank prefer rising rates). Interesting. Perhaps that varies even over time, e.g., with the preference for holding 30 year Treasuries.

rising inflation gashes the value of of your loans, particularly fixed rate mortgages, which got so many thrifts in trouble in the 1980s.

I think you mean the trouble grew manifest in the 1980s (while Congress ignored Edwin Gray of the Federal Home Loan Bank Board). Escalating inflation was a phenomenon from 1965 to 1974 and from 1977 to 1980, not afterward.

There are some banks which benefit from rising rates("asset sensitive") and some which benefit from falling rates ("liability sensitive"). It depends on the duration and structure of your assets and liabilities. If you have a lot of commercial and industrial loans, which reprice quickly on LIBOR, and a lot of sticky demand deposits, which will reprice slowly or not at all, you want rising rates. (JPM is an example of this type of bank). If you mostly have residential mortgages, which are long-term fixed rate, and borrow through CDs which reprice relatively quickly, you don't want rates to rise at all. (Your local thrift might be an example of this). There are a lot of banks in between. Therefore you can't look at banks in aggregate and conclude that rising rates are definitely good and definitely bad. Higher rates are probably eventually good for all banks but a liability sensitive bank would want rates to rise very very slowly. I would say it's probably true that more banks are asset sensitive but the ZIRP environment probably makes more banks more liability sensitive than is usually the case.

I think you have to be very careful when you say "high rates are good" or "rising rates are good". DeLong is right that if inflation eventually got to 4%, NIMs for all banks would eventually be higher, but if you raised the inflation target to 4% tomorrow, banks with huge books of 3.5% 30 year mortgages would be in big trouble, so of course banks would be against such a policy.

This is a political economy question. If the results of Fed policy are so ambiguous, and I think it is,why the unanimity of media macro that interest rates have rise NOW if not last June?


It's interesting that Trump's chance of winning if nominated is higher than Bush's or Rubio's. I would think Trumps relative appeal to moderates and independents would be far lower.

What's even more interesting (not from a political standpoint, but from a rationality standpoint) is that Martin O'Malley is given a higher chance to win the general election than he is to win the nomination. His conditional probability stands at 180%.

Bit of a shameless self-promo, but #2 looks an awful lot like a copy of a post I made last week:

As I wrote in my post, I have no proof, but I'd guess that longshot candidates have higher electability numbers because they'll only get nominated if something crazy happens in their favor, which would also benefit them in the general. This has been mildly borne out by the case of Scott Walker, who had dismally low electability rating when he was one of the "Republican Big 3" with Bush and Rubio, but now that Walker is a longshot candidate his electability number has gone up quite a bit

Right, I get that, but that's not what I was commenting on. What the betting markets are saying is mathematically impossible (unless they think that there's a real chance Martin O'Malley is going to run as an independent and win) because they think that there is a 0.2% chance that he is going to be President, but only a 0.1% chance that he'll win the Democratic nomination in the first place.

When the numbers are that low, the noise drowns out the signal.

I'd guess the O'Malley results aren't particularly meaningful because trading commissions and fees tend to make betting on extreme longshots unattractive

And FWIW right now on Betfair if you want to bet on O'Malley, you'll get slightly longer odds on the presidency (1/200 = 0.5%) than on the the nomination (1/160 = 0.625%), so the situation has at least flipped back to "mathematically coherent"

Such results do seem to point at a flawed algorithm.

"I would think Trumps relative appeal to moderates and independents would be far lower."

This does not surprise me at all since Trump was a Democrat for many years.

Krugman seems a little confused. I work for an insurance company and I can tell you that low interest rates have been fantastic for profits. Moreover, we get squeezed in our note issuance when there is a small spread between treasuries and the corporate rates. The relationship between low rates and low spreads might be real (I haven't investigated this), but it's the low spreads that hurt "profits" (a word which I utterly despise), not the low rates. In other words, it's the spreads stoopid.

#5: The most startling trend in economics publishing, to me, is that top journals now publish a lot of papers that have nothing to do with economics at all. (Unless economics is defined as the use of regression analysis.)

#3: Wearable GPS is standard in the Australian Football League. I don't know how that compares to other sports - I guess I had assumed that richer sports would be far ahead of our game.

You can see the outline of the device in #14's guernsey in this photo:

Getting a grip on the real state of China's economy is hard. Telcom services are another data point. Since it isn't clear if we are getting revenue or volume numbers, the meaning of the data is ambiguous.

Some time ago, I checked China electric power data. Power production shows ongoing, relatively fast growth. Of course, the power production data may not be accurate.

For better or worse (worse) the China bears appear to be overly pessimistic.

It's really sad that Romney did not try again. I think he was the best GOP candidate in years.

Indifferent candidate, but very accomplished and with executive experience in business and government. Not since Gen. Eisenhower has there been any candidate better prepared for the job.

The most notable thing about the Chicago Fed article is just how tiny the effects are. In fact, it would have better illustrated the results in the article if they just showed zeros in all the graphs.

If you look at the graph on the first page, it shows that for banks with more than $10 billion in assets (which account for over 80% of banking industry assets overall), a 1 percentage point increase in the three-month Treasury bill interest rate results in only a 0.3 basis point increase in the net interest margin, or a 0.003 percentage point increase. According to this data, then, the *direct* impact on banks' margins is so completely insignificant that it can - in practical terms - be disregarded.

This result is almost so extreme that I have a tough time believing it - I wonder if the study isn't accounting sufficiently for lags, etc. But in general, it does ring true: if you just eyeball a graph of the fed funds rate against the banks' net interest margin, it's hard to discern much of any relationship. See, for instance,

My guess is that banks structure their balance sheets so that their returns look roughly stable in an accounting sense, even if they're not as stable in a true economic sense (which is hard to pin down anyway). There seem to be two offsetting elements here:

(1) Interest rates paid on deposits don't increase one-for-one with market interest rates for comparable maturities. This is especially true for checking accounts and small-time savings and money market deposit accounts (and less true for CDs), and it means that banks' implicit margin on their liability side increases with interest rates.

(2) Banks overall have a maturity mismatch, with longer maturities on the asset side. Many of these assets are held to maturity rather than marked to market on the balance sheet; within the lifetime of a given long-term asset, therefore, a bank will have an artificially high net interest "margin" when short rates are low and an artificially low "margin" when short rates are high, even when the original yield on the long-term asset fully anticipated this progression of short rates.

Banks use some ad-hoc mix of (1) and (2) to keep their profits *looking* relatively stable, even if this is more accounting than anything else.

One question is why, exactly, banks can get away with the sluggish response of deposit interest rates in (1) -- in a free market for deposits, why can they charge 4% below the market interest rate some times and 0% other times, even when the true cost of providing the intermediation is probably constant at around 2%? The only answer that I can see is adjustment costs on the part of consumers: the vast majority don't shift banks in response to short and medium-term fluctuations in interest rates, especially for accounts that are frequently used for transactions and transfers (like checking accounts, most notorious for their less than 1-for-1 response of yields). When customers are sufficiently sticky, the perceived accounting benefits from using (1) to offset (2) drive banks to keep deposit rates relatively stable.

I wonder whether the rise of internet banking will upend this arrangement.

#1...Paying into a govt run health plan only restricts your liberty if you have to use it. Otherwise, you're simply complaining about paying taxes for things you'd rather not, but we're all in that boat. Also, arguing that taxes inherently limit liberty ends up as an argument for the view that having more money gives a person more liberty, and a lot of people won't agree to having less liberty than others, if you put in those terms.

Also, I don't like paying taxes, but I'm generally left with money after paying my taxes. On the other hand, if you make something I value illegal, then my liberty really is limited, since I can't do it at all without going to jail. As for the GOP, hardly a day goes by without a GOP politician proposing to severely limit liberty.

Finally, if you argue we need to spend and spend more on the military to preserve liberty, a pragmatic argument, you could as well argue on those grounds for social services.

The only way to have smaller govt is to grow the economy in such a way that people don't feel they need it.

As for the GOP, hardly a day goes by without a GOP politician proposing to severely limit liberty.

Spicoli's liberty. The threat to the liberty of the adult population comes from the legal system.

#1 I feel neither Boudreaux nor Kling state the real objection to Rodrik's piece, which is that his analysis is highly nationalist. It is true that under certain conditions, particularly incomplete trade liberalization and market inefficiencies, you will have trade where one party does lose out, perhaps even one of the parties within the trade itself. However Rodrik is still not averse to these trades occurring intranationally, basically because he understands that these trades will still be *efficient*. Thus, the workers in the sector losing out can migrate to the winning sector, or be given a Kaldor-Hicks compensation from the gains of that sector. An overarching domestic policy regime makes this possible. But then it follows that his main objection to foreign trade is that he doesn't like any trades where his "home" trading partner might lose out, even though from a global perspective the result is likely still efficient, as he can't enforce the redistribution required to ensure that everyone does win. Now add to this the fact that such trades will generally be between the US and a country poorer than itself, so improving global equality as well as efficiency (though hurting domestic equality). But because he is nationalist, the latter is most salient to him, and governs his policy advice. This is also why he brings up the optimum tariff argument in the linked post - of course this is just the advice that a big trader like the US behave monopolistically (ignoring the threat of retaliation), which of course will be bad for global efficiency but lead to profit for the US, if that's what you care about most. And yet the "textbook" trade policy advice generally implicitly speaks from this perspective, so his college-trade-professor demeanor with which he states his objections shares that bias.

Comments for this post are closed