Relevant and even prescient commentary on news, politics and the economy.

I actually disagree with Paul Krugman for once

This is an exiting day. I disagree with something Paul Krugman wrote.

In 2017, private insurance paid about a third of America’s medical bills — $1.2 trillion, or 6 percent of GDP. Having the government pay those bills directly, without a revenue offset, would therefore be a spending increase — a fiscal stimulus — of 6 percent of GDP.

Suppose — as MMTers tend to assume — that interest rates nonetheless didn’t rise. Then this stimulus would have a multiplier effect, probably raising GDP, other things equal, by 9 percent.

I have 2 objections. First the replacement of private insurance with debt financed single payer is effectively a tax cut not a spending increase (as indicated by the phrase “those bills”) as such, the direct effect on demand is less than 6% of GDP. Krugman likes to do two kinds of analysis IS-LM and New Keynesian.

In a standard new Keynesian model, the shift would have no effect on demand — ultra rational consumers would assume that they would have to pay the public debt eventually, so they would save the money that isn’t being paid as insurance premiums (which would presumably be paid as salaries instead).

In an IS-LM model, the incrase would be 6% times the marginal propensity to consumer. Going full Hicks (the orignal IS-LM model) that is 1-1/(the multiplier) = 1/3. To consistently apply the original IS-lm model, Krugman should calculate (multiplier -1)(the tax cut) = 3% of GDP.

Second Krugman writes “a multiplier effect” when he means “a multiplier greater than one”. I hate that. It does not follow from the definition of “to multiply” If I have to multiply a by something to get b, that doesn’t mean b is greater than a. Mutiplying by 0.9 is multiplying.

I insist on this, because anti Keynesians often play the 1=0 trick. When fiscal stimulus is proposed, they claim to prove that the multiplier is zero. When data is analyzed they claim to have been proven right, because there isn’t proof that the multiplier is greater than 1.

But the important point is that the correct calculation implies a 3% increase in GDP. Okun’s law (click the link) implies a 1.5% reduction in unemployment to 2.5% which would probably scare the Fed into raising interest rates, but which is a lot more possible than -0.5% as calculated by Krugman.

Now I don’t think that the marginal propensity to consume is 1/3, but that means that Krugman and I have to explain why estimate multipliers are only 1.5 and not much higher. I suspect this explanation would imply that MMTers predict a negative unemployment rate, so I suspect thaat, in the end, I would agree with Krugman’s conclusion.

But I find his calculation suspect.

Comments (8) | |

The Ethics of Clinical Trials

In a clinical trial the therapy is decided by a pseudo random number generator. How can this be ethical ? People are treated differently for no reason related to different interests different values and priorities or even different merit (assuming merit can differ).

There is a utilitarian rational for clinical trials. Through such trials doctors learn, and that knowledge is useful to future patients. But this rationale is utterly rejected as ethically unacceptable, because it was used to justify depraved experiments.

I think the current discussion of the ethics of clinical trials is based on a mixture which is partly consequentialist and partly deontological, and that it is incoherent, because people feel the need to claim it is totally both, while the two are inevitably in conflict.

So it is asserted that physicians must act in the interest of the patient – each and every patient. It is also argued that clinical trials are morally acceptable. This does not make sense.

It is only possible if the expected welfare of the patients is identical under the two treatments over which one randomizes. Any difference, no matter how tiny, in expected welfare would compel the use of only the current standard therapy, or of only the new experimental therapy.

I think the failed effort to avoid this is to reject the concept of expected welfare. It is argued that it is OK to do one or the other because one does not know which is better for the patient.

It would be OK if one were to say all probabilities must be rounded to 0, 1 or 0.5 so we don’t know means each is exactly equally likely. However, this approach would make life strange and brief. In particular it would rule out general anesthesia for any procedure not necessary to save a life. The chance of death is very low but demonstrably not zero. Don’t operate unless you would operate with a 50% chance of killing the patient would rule out almost all surgery. We must make choices under uncertainty and can’t pretend that all uncertainty is the same and survive for long.

Consider 2 examples:

1. There are 2 treatments, and, with best estimates, with treatment A the probability that the patient lives is 50% and with treatment B the probability is 30%.

2. There are 2 treatments, and, with best estimates, with treatment C the probability that the patient lives is 50% and with treatment D the probability is 30%.

According to current medical ethics, one must provide treatment A not treatment B but one may chose treatment D or treatment C. This always is based on the assertion that the interests of the patient is all that matters. Yet I have assumed that, for the patient, the two pairs of choices are identical. This can’t make sense.

In the first case there is an unobservable difference between patients of type 1 or type 2 where if they are type 2, then treatment A kills them on the spot. 10% of people are of type 2 (as learned from decades of painful experience). If someone is of type 1, their chance of surviving with treatment A is 5/9. In contrast with treatment B all have a 30% chance of living. With decades of painful experience it is known with essentially complete certainty that the probabilities are 50% and 30%.

In the second case, there aren’t two types, but the evidence on treatment C is preliminary based on a small (phase II) trial. The fraction who survived in the trial was 50% but the 95% confidence interval is 20% to 80%. The null that the true chance is 30% is not rejected at standard confidence intervals. By standard reasoning it is time for a phase III trial with randomization.

In each case, we know that giving A not B might cause a patient to die who would otherwise live and our best estimate of the probabilities of survival are higher with A than with B and higher with C than with D.

I think the difference is that one learns something by randomizing and giving half of the patients D and that this outweighs the expected deaths due to the randomization.

I think it is possible to believe people have a right to care, and also conduct randomized trials, if one says there must be a standard of care, and all people have right to that. That one may deviate if the weight of evidence suggests that an experimental therapy is better, but that such deviation is a matter of utilitarian total expected welfare maximization not individual rights which trump average interests.

But it is not easy or comfortable to believe this, so I think that doctors have decided to rely on statistics but reject the very concept of probability. The logical inconsistency might cause some discomfort. It would cause more if the concept of probability weren’t so utterly alien to normal human thought. But in any case the tension between believing in rights and believing those rights don’t always trump utilitarian calculations clearly causes more discomfort.

Comments (2) | |

A Facebook Experiment

Solid social science on the opinion pages (needless to say news reporters consider interest in randomized controlled experiments to be opinion.

Christian Caryl explains how it is possible to determine the effect of the Russian influence campaign on the 2016 presidential election.

Sinan Aral, a professor at the Massachusetts Institute of Technology [skip] says he and his colleagues want to study the Russian influence campaign in precisely this geographical context. The MIT scholars have developed a robust methodology for assessing how social media campaigns influence the behavior of their targets — and now they want to bring it to bear on the Russian meddling in 2016 [skip]

“For example, Facebook and Twitter constantly test new variations on their feed ranking algorithms, which cause people to be exposed to varying levels of different types of content,” they write. “One underpublicized A/B test run by Facebook during the 2012 U.S. presidential election caused users to be exposed to more ‘hard news’ from established sources, with effects on political knowledge, preferences, and voter turnout.” Given access to adequate data, the researchers claim they can estimate the impact of the Russian influence campaign in Michigan, Wisconsin, Pennsylvania and Florida “with 95% to 99% confidence.”

Facebook performed the necessary experiment, because they perform experiments all the time aiming to maximize user engagement. These are genuine randomized controlled experiments (because Facebook’s profits are on the line). The names of people exposed to more or less hard news can be compared with the lists of people who actually voted (which are public) only if facebook is forced to cease to protect their privacy (which really means to protect Facebook from proof that they let the Russians trick Americans into electing Trump).

I don’t think the researchers will ever get access to “adequate data” but I do think it is worth fighting for such access.

update: this might not have been clear in the original post. The idea is to use which of the 2 algorithms was used as an instrument for exposure to Russian propaganda. So it is engagement with known russian propaganda regressed on the algorithm used to check the association of (engagement with known Russian propaganda)(demographic characteristics typical of Democrats) and turnout (in particular of African .

To be crude (and explicit) the idea is that African Americans with the hard news algorithm interacted less with Russian propaganda and were more likely to vote. Or not (good experiments are ones where one doesn’t know the result before analysing the data).

Comments (4) | |

How did they get so rich ?

I hope and trust that this will be an amusing display of my ignorance. I don’t hope to reach David Graeber’s level

David Graeber: Apple Computers is a famous example: it was founded by (mostly Republican) computer engineers who broke from IBM in Silicon Valley in the 1980s, forming little democratic circles of twenty to forty people with their laptops in each other’s garages…

1. How did Jeff Bezos get so rich ? His wasn’t a subtle idea.
The first point is that Amazon went for years without turning a profit. I think his strategy was partly based on new entrant predatory pricing. I would guess that are many B league Bezoses whose firms went bankrupt. In the field of innovative low margin retail, survivor bias is a bitch.

But also, books were a good place to start. There are lots of different books. They are durable. They aren’t so heavy (even pre-kindle).

Then there is big box one stop shopping.

2. Yeah what about the Waltons ? At first (in the 80s I think) I was very puzzled to read the name Sam Walton on a list of the super rich, because I had never seen a Walmart. In fact, that was when I first read “Walmart”. Retail is a very competitive low margin sector. How could anyone become super rich in discount retail ??? My guess is that he was the first to realize how much big cars had changed the game. The long cars of pre-1973 had big trunks, but I guess they just aren’t in the same league as pickups, SUVs and minivans. It is notable that Walmart started in the huge vehicle belt (which I think has a lot to do with belt size if you get my drift).

3. Brin, Page and Google. OK look the original product is actually excellent. Also giving stuff away (including 1 gigabyte e-mail boxes) was smart.

Comments (5) | |

The Buffett Buffer

This is an almost semi serious proposal suggesting payday lenders could get good publicity.

There are many entities with plenty of spare cash. The US Treasury isn’t one of them. I think that the good publicity gained by offering zero interest loans to unpaid federal employees is worth the cost. They are good credit risks because they will get paid (without interest) eventually.

I’d say some entity with spare liquidity could help the country and win praise. I call my proposal the Buffett buffer. I think the challenge is largely one of automatic underwriting. It would be necessary to extend the offer only to people who are employed by affected departments. I don’t think this is hard — the web pages are still up, so it is possible to search them. The skeleton staff of non furloughed workers could make employment information public at the request of employees (including their un-paid selves).

Actually since the payoff is in popularity, I think this might be the Bloomberg boom not the Buffett Buffer.

Why not ?

Comments (1) | |

Romer & Romer on Taxes

Given the debate about returning to 60s level top marginal tax rate of 70% amazingly re-opened by Alexandria Ocasio-Cortez, I decided to actually read the Romer and Romer paper (pdf warning) which includes evidence suggesting an even higher rate is optimal. It is a masterpiece, which I won’t try to summarize. Read it.

I do however, want to grind a very old ax related to “Schlock Economics”. I am thinking of the time that Robert Lucas totally humiliated himself while accuding Christine Romer of Schlock Economics. He demonstrated that he had managed to forget the IS-LM model (honestly he might actually have failed my undergraduate intro-macro course which is a major major accomplishment few have achieved).

Old ax grinding after the jump

Comments (0) | |

Optimal Taxation of Capital Income 2019 (let them Bern).

I wrote a post about optimal taxation of capital income which (the web is sometimes wonderful) was made legible by the blessed [person who choses to remain anonymous].

But that was back in Obama center left 2008. I want to update given what I learned since then and given the appearance of socialist US citizens.

First, what I should have known already is that the standard Judd 85/86 result that the optimal rate of taxation of capital income goes to zero as time goes to infinity is what mathematicians call a boo boo (oopsie). The asserted theorem is false as explained by Ludwig Straub and Ivan Werning.

This is an interesting event in the history of thought and the sociology of economics — a standard mathematical result which is simply wrong. It is especially interesting as the proof that Judd made a whoopsie was published years ago, yet the false alleged result survives. One might almost suspect that ideology or class interest is involved.

The key issue is that Judd considered tax rates which change over time and their incentive effects and then casually assumed that the public sector budget is always balanced. I guess he guessed this was OK because of Ricardian equivalence which says that. given a long list of implausible assumptions, the timing of *lump sum* taxes doesn’t matter, so the timing of taxes only matters because of incentive effects.

In fact Judd’s alleged proof is completely invalid. It is simply a math mistake.

The model
There are 2 groups workers and investors. The workers consume all of their income which consists of a wage and, possibly, a subsidy from the state. Investors have capital income — interest after tax A_tf'(K_t)-tau_tK where tau_t is the rate of taxation of capital, A_t is their wealth and K_t is total capital (these must be equal under Judd’s assumption that the state neither borrows nor accumulates a sovereign wealth fund).

Investors maximize an intertemporal utility function with rate of imaptience rho. The claim is that if the state wishes to maximize a weighted average of workers’ instantaneous utility and investors’ instantaneous utility and also has rate of time preference rho, then Tau_t goes to zero as t goes to infinity.

Now first note that even if Judd were right it would tell us nothing about what taxes will be optimal for the next million years. Oddly, many people some of whom are economists (one of whom Edward Prescott has won the Nobel memorial prize in economics) conclude that taxes on capital income should be cut to zero right now.

Second allow the state to accumulate wealth and consider the simplest case in which investors maximize the discounted stream of the logarithm of their consumption. This means that they consume (rho)A_t no matter what Tau_t is. Assume that tau_t can’t be greater than some limit taumax or the state will grab capital instantly which is, in effect, a lump sum tax and doesn’t distort.

In this case, Tau_t does go to zero, because the state accumulates until it’s income covers all its expenses plus whatever subsidy it chooses to pay workers and the distribution of income is exactly that which it finds optimal and then ceases to tax as there is no reason to tax anyone. Note that in this case there is no trade off between efficiency and desired redistribution — the distribution converges to that desired as if there were no problems with incentives. This is roughly the opposite of the standard interpretation. In the long run, the distribution of income is exactly as desired. There is no more taxation because there is no more reason to tax.

(More generally if the elasticity of substitution is less than one in the model (as it is in the data) the state will redistribute more until the investors are relatively poorer than is optimal. This is because the income effect of the tax is greater than the substitution effect, so high taxes on capital income promote saving. But I want to mainly stick with logarithmic utility).

Now consider an extreme case in which the state cares only about the welfare of workers. Judd claims the result holds even in this case. He is wrong even if the state is allowed to accumulate a sovereign wealth fund. In that case, there is a loss due to investors consumption equal to (rho)A_t and the state aims to mimimize A_t. If there is an upper limit on Tau, then Tau_t is always at this limit. The optimal policy is to tax capital income at the maximum rate allowed forever.

Now consider Judd’s assumption that the state can’t accumulate a sovereign wealth fund. The fact that it must leave the wealth in the hands of investors who consume it at rate rho means the optimal steady state K=K* is what it would be if there were depreciation at rate rho. This means that f'(K*)=2rho.
For investors to choose this steady state, it must be that the after tax return on capital (f'(K*) – tau) is equal to rho so 2rho -tau = rho and tau=rho.

So where did Judd go wrong ? His alleged proof included the assumption that the economy would converge to a steady state. He assumed that A = K, but also considered only the social budget constraint and concluded that, in the optimal steady state f'(K) = rho (as would be true if the state had access to optimal non distortionary taxation or if it could accumulate a sovereign wealth fund). But the restriction A=K is binding & it has a non zero shadow price. That shadow price is not constant even if all observables K, A, w, R consumption of capitalists and consumption of workers are constant.

Given the problem as stated, the economy can’t reach a steady state. The gain to the social planner of being able to accumulate wealth becomes constant. It’s current value grows at rate rho.

The impressive thing is that the alleged result is still accepted even though the proof that it is a math mistake was published over a decade ago.

Comments (0) | |

The Price of Naltrexone

In The New York Times Abby Goodnough wrote
” she got a Vivitrol (naltrexone) shot but it was so expensive — her co-payment was $600 — that she never got another” !!!

This is insane. Naltrexone is an opioid antagonist. It prevents opioids from causing a high (and relieving pain and suppressing coughing and breathing). In no way is it conceivably a drug of abuse. But opioid addicts who wish to cut off all effects of opioids have to pay for their Naltrexone.

Also (as explained in the excellent article) some of the same people who oppose the use of methadone and buprenorphine oppose naltrexone too. I have never understood their logic. I am sure it is based on a moralistic belief that there are no simple easy solutions. It isn’t even “no pain no gain” as cold turkey withdrawal while using naltexone is just as horrible as any other cold turkey withdrawal. Pointless speculation after the jummp.

But for now two practical proposals. Everyone who wants naltrexone for any reason should be given naltrexone (given no co-pay). I think this is obvious. Now somehow a drug which has been around practically forever is expensive, but the cost of paying off the pharmaceutical company whatever they demand for such a program (which will be great for them) is trivial compared to the costs of the opioid epidemic.

I should have provided a link to the Wiki on Naltrexone. Note the cost (retail) of oral Naltrexone is $0.74 a day — providing one a day to every addict and anyone who wanted to pretend to be an addict would cost hundreds of millions a year. This is a completely insignificant sum for the US government, so it should be done immediately. Delayed release Naltrexone is expensive (prescribing it with a $600 copay is bad practice of medicine). Here a technological improvement has made it possible for doctors to give the patients a better, but expensive option, which they don’t take.

I also have an impractical proposal that Naltrexone should be available over the counter — it can’t be abused and the reported side effects are the reported symptoms of being a person. However, I know this proposal is impractical.

My second practical proposal is phased drug assisted therapy. I think it should be
1) whatever you want for a week provided you don’t want a lethal dose (you want heroin — here’s your heorin)
2) second week whatever you want provided you take your methadone under our supervision. All the heroin you want will be none (it doesn’t do anything for someone full of methadone).
3) third week, 50% methadone 50% buprenorphine.
4) fourth week buprenorhine
5) fifth week 50% buprenorphine 50% naltrexone
6) 6th week through death do us part naltrexone.

Why not ?

Comments (2) | |

Twelve Dimensional Chess

Barack Obama just won Gallup’s man most admired by US adults poll for the 11th straight time.

Also, in spite of the Republicans’ determined efforts including elimination of the mandate, Healthcare.gov signups are only 4% fewer than last year’s.

Finally, I am thinking about “eleven dimensional chess”. This was a joke about Obama and Obamaniacs who ascribed his amazing luck to brilliantly subtle strategy. To be honest I was thinking of Hillary Clinton (second to Michelle Obama after 16 first place finishes) as someone who managed eleven dimensional fools mate.

And I recall that the original 11 dimensional chessplaying was cleverly promising health care reform could work without an individual mandate. A merely sly politician would promise this, because he knew that reform was popular, mandates are not popular, and it is better to break a promise after being elected than never to be elected at all. But, it was argued, Obama campaigned in the obvious way on an obvious lie exactly, because he knew that reform without a mandate would be a catastrophe and especially catastrophic for the insurance companies. By making a promise which shouldn’t be kept, he scared them and their tame senator Max Baucus into supported reform if and only if there were a mandate. He also 11 dimensionally managed to cede control to the Senate as is necessary to flatter senators (and avoid the blame for the necessary tradeoffs). The Clinton’s sacrificed their king (and hurt the electoral chances of their queen) by setting up a committee and keeping Senators waiting for months.

But wait. Didn’t I just mention that health care reform is actually functioning without a mandate ? Is it possible that it can work with subsidies which prevent an adverse selection death spiral ? As Obama claimed in t008.

Now that, that is 12 dimensional chess. He was so brilliant that he managed to promise something which was possible (and hasn’t hurt the insurance companies) yet terrified them into supporting health care reform. In this way he 12 dimensionally convinced Republicans that they could destroy healthcare reform by eliminating the mandate, so they did, and we ended up with the policy he most slyly proposed.

Either he is a true genius of 12 dimensional chess or he is very very lucky.

Comments (8) | |

How much did that bailout cost ?

11 years after the huge financial rescue operation, procrastinators look at the cost to the Treasury. The numbers are gigantic. Also the cost was negative. Saving the financial system and preventing a second great depression was, I think, the most profitable trade in human history by far. Crude accounting suggests this, but there are two relatively sophisticated arguments that the rescue didn’t yield profits and was, in fact, costly.

First the notorious TARP was a small part of the rescue. Fed purchases of risky securities at prices far higher than anyone else was willing to pay dwarfed TARP. Also the separate rescue of Fannie Mae and Freddie Mac was very large compared to TARP. This means that the profit earned on TARP (which is small only compared to the huge amount of wealth at risk) is misleading. In fact, the Treasury made a profit on TARP even including the too small program to help mortgagers (HAMP) which was a gift not a loan and the cost of saving the US automobile industry. But focusing on that implies missing the much bigger picture involving federal reserve banks and the government sponsored entities which have become (again) government owned entities. If one considers them, one sees that the rescue was not just profitable, but the most profitable deal ever.

Here is the cash flow. It is huge and the cash is flowing in to the Treasury.

These are large numbers. The dividends largely resulting from the rescue are roughly 10% as large as total dividends paid by corporations.*

The second argument is that crude accounting is vulnerable to extending and pretending — non performing assets appear on balance sheets as if they were worth their face value. The Federal Reserve Banks do not mark their assets to market. It was arguable that their accounts were deceptive and that the profits they weree required to hand over to the Treasury would be more than balanced by losses which the Treasury would have to bear. Now there has never been final proof of anything ever, but I think it is safe to say that this reasonable argument was incorrect. The rescue might end up costing the Treasury if there is a financial crisis which dwarfs 2008. The assets of the Fed, Frannie and Freddie might become worthless, for example if a very large meteor hits the earth. But that’s not the way to bet.

Also, I told you so on September 18 2010

Comments (10) | |