Pissing Down Your Leg

Thoughts on Economics and Economic Policy

The Power of Monopsony

leave a comment »

I work at a public university. Like many such institutions, we have been dealing with reduced support from the state. In fact, the percentage of the school’s budget that is funded directly by the state has fallen from 50-60% in the 1970s to about 10% (I think it’s less than 10% currently). The school has had to increasingly rely on tuition hikes in order to fund itself. This, of course, is increasingly unpopular with students. No longer can they attend their home state public university at a reasonable cost. In-state tuition is about $12,500 before room and board, about three times more expensive in real terms than in the mid-1980s. Out-of-state tuition, on which the university relies heavily has also seen similar increases. So the only public university in the state receives about 2.4% of the state’s total expenditures.

This seems like a bad idea. In a world in which economic success increasingly relies on completing some level of higher education, the state is not investing in the human capital it will need to grow in the future. Parents and students are upset. Professors and administrators are upset. This doesn’t seem like a political equilibrium. And yet the trend continues. The question, then, is why?

I believe that the answer is the rising cost of health care and specifically the larger and larger percentage of the budget taken up by Medicaid. Medicaid expenditures, of which the state pays slightly less than half of the total (the Fed picking up the rest), now account for 30% of the state’s total budget. This, of course, is up from 0% in the mid-1960s. As Medicaid eats up a larger and larger piece of the budget pie, everything else has to shrink.

So what do we do? I suppose one solution would be to leave our poorest and most vulnerable citizens without access to health care. But I don’t believe that would be the best solution. Instead, we should look at the best way to control our health care costs. And this is where the power of monopsony comes in. A monopsony is like the opposite of a monopoly. In a monopoly, there is only one seller. That seller can maximize its profit by increasing its prices and reducing its output compared to what would exist if there was more competition. In a market with a monopsony, there is only one buyer. That buyer can reduce the sellers’ profits by paying less than they would have to in a more competitive market.

Now normally, I wouldn’t encourage either monopoly and monopsony in any market. But the health insurance and health care markets are not like any market. They are fraught with problems of adverse selection, moral hazard, and information asymmetries. These markets, left on their own, will fail to achieve efficiency 10 times out of 10. That’s where monopsony, in the form of a national, single-payer health plan comes in.

With single-payer health care, everyone is enrolled and pays for it through a tax (similar to Medicare). Out goes adverse selection and moral hazard. The government is in the position to provide information on the efficacy of treatments and only pay for those that work. And the single plan is in a position to bargain prices with doctors, hospitals, and pharmaceutical companies to keep costs low.

The United States spends more than twice as much per person on health care than the median OECD country. And it does so without getting better results. If we adopted a single-payer plan, the potential savings would be measured in the trillions of dollars. We could increase funding to public education (that’s where we started) and infrastructure and everything else that’s being starved. What’s not to like?

Always remember that one person’s spending is another person’s income. Spending $1 trillion less on health care each year means $1 trillion less in income for doctors, nurses, hospital staff, and pharmaceutical workers. This would have to be a gradual process just so we didn’t throw the economy into recession. But these groups will fight tooth and nail to protect their incomes. Bargains will have to be made. I would start with paying off education loans for health care workers. It may also be a good idea to throw more federal research money to the pharmaceutical companies.

But it doesn’t make sense to allow one industry, even one as important as health care, to eat up more and more of our resources if we don’t have to. The power of monopsony would allow the federal and state governments to spend less and less on Medicare, Medicaid, and other health programs, opening up money for investing in the areas that strengthen our economy for the future. What kind of suckers are we that we don’t do this?

Written by Liam C Malloy

June 1, 2012 at 11:13 am

Sovereign Debt and Eurozone Exit

leave a comment »

So my previous post ignored the 800-lb gorilla in the room, namely sovereign debt. The goal of exiting the eurozone would be to make your economy more successful and allow monetary policy that is appropriate for your national economy. If successful, this will lead to reduced unemployment and higher tax revenue. It will not solve chronic deficits.

So I believe that sovereign debt should be converted from euros to the new national currency at the shadow exchange rate. This would avoid a European-wide banking crisis as investors should be indifferent between the old debt in euros and the new debt in lira (or pesetas or drachmas). But it wouldn’t solve the Greek problem of continuous deficits due to overspending and poor tax collection. I don’t believe there is any monetary solution to that issue. It should, however, greatly help countries like Spain and Italy which have not been financially profligate.

Written by Liam C Malloy

May 11, 2012 at 9:06 am

Posted in Eurozone

How to Break Up the Euro

leave a comment »

Let me start by saying that this is completely outside of my area of expertise (assuming I have one). But sometimes an outsider’s perspective can be useful. And because a breakup of the eurozone seems increasingly likely, here is my suggestion on the best way to accomplish that.

The problem, as I understand it, is that euros are completely mobile within the eurozone so that if Greece (or Spain or Portugal or Italy or Ireland) looks like it’s about to leave the currency in favor of the drachma, everyone will take their euros out of Greece and put them in Germany. Then as the drachma depreciates against the euro, they can exchange their euros for drachmas without losing their real purchasing power (and even increasing it within Greece). This will cause a banking crisis making the exit extremely painful.

So we have two related problems with which we have to deal. First, how do we keep the euros from flying out of the countries that are most likely to leave the eurozone and adopt a new national currency. And second, how do we price those new currencies against the euro. The problem is that we’ve been assuming that we can’t do the second until the countries have left the euro, making a solution to the first problem seem impossible.

The solution requires that we price the potential new currencies (against the euro) before any countries actually leave. If the new drachma (or peseta or lira or Irish pound) is priced appropriately, there will be no desire to move euros around. For example, let’s say that in order to get Greek prices back in line with the rest of Europe, the new drachma would have to depreciate 100% so that the exchange rate would be two drachmas per euro. If that’s the “correct” real exchange rate, so that the purchasing power of two drachmas in Athens is the same as one euro in Frankfurt, then people will be indifferent between having two drachmas and one euro. If Greek depositors know that they will get two drachmas for each euro in their account, then they will have no incentive to move those deposits to Frankfurt before conversion.

The goal of this plan is to create a shadow exchange rate for each country within the eurozone (although it’s especially important for those most likely to leave). This way you know that you’ll get two drachmas for every euro or 1.3 pesetas or 1.5 Irish pounds. This would keep the capital in the troubled countries because depositors and investors would know that their value was safe.

How do we do this? We need to create a market for these nonexistent currencies. While I’m not 100% sure about the best way to do this, I do have a suggestion. I suggest that we create dual-track eurobonds. The first track would be straightforward bonds of varying durations. The interest rates would be determined by auction as in most bond issues (again, this isn’t my field, but this is my understanding of how things work). The second track would take the interest rates from the first track, but would introduce the possibility of being paid in euros or a new national currency. The exchange rate would be determined by auction in the same way that the interest rates were determined for the standard bonds.

For example, let’s say the regular 1-year eurobond is paying 2% interest. If you buy a €1,000 Greek 1-year eurobond at an exchange of 2 drachmas to the euro, there are two payout options. First, if Greece stays with the euro you will get your €1,000 plus 2%, or €1,020. It will behave just like the standard bond. However, if during that year Greece exits the eurozone, you would get paid in drachmas. In this case your principal would be ₯2,000 and with 2% interest you would receive ₯2,040 (is that the right symbol for the drachma?).

These shadow exchange rates would be public information so that every depositor and investor would know how many drachmas she would receive for every euro in a Greek bank or in exchange for euro currency in Greece. So how would an exit from the eurozone work? First these bonds would be set up, ideally for each country in the eurozone, but certainly for those countries most likely to leave (Greece, Spain, Portugal, Ireland, Italy). The bonds will then tell us how much depreciation each currency would need in order to keep investors indifferent between holding euros and drachmas. It would then be up to each national government to decide whether or not to stay in the eurozone or adopt a national currency.

If the country decides to leave the euro and adopt the national currency, this is what I assume would happen: The exchange rate is determined by the national eurobonds. On the day of the conversion, bank deposits and other financial investments (including the national eurobonds) within the country are switched automatically from euros to the new currency. If the exchange rate is 2 drachmas per euro and you had €5,000 in your account, you now have ₯10,000. All prices within the country (including wages) are switched one-to-one to the new currency. That is, if you were earning €30,000 in Greece as a school teacher, you are now earning ₯30,000. If a bottle of Greek retsina cost €10, it now costs ₯10. So the real wage for domestically produced goods stays exactly the same. But of course the cost of imports will rise and the cost of exports will fall (remember, that’s why we’re doing this in the first place). A BMW that used to cost €20,000, or 2/3 your annual salary will now cost ₯40,000, or 4/3 your annual salary. On the other hand, that €10 bottle of retsina will now only cost €5 in Frankfurt. And Germans will find a vacation in Greece to be much cheaper than before. This should boost Greek exports and employment.

The whole key to this plan is that we don’t know what the exchange rate should be. We know it isn’t 1:1. That’s why we’re having this problem in the first place. But we won’t be able to figure it out. We need to use the market, investors with real money on the line, to figure out the appropriate rate. Maybe it’s 1.2 drachmas per euro, maybe it’s 1.5. I have no idea. But if investors are willing to buy the national eurobonds, this means they are indifferent between euros and the new currency at that rate.

Will people be able to game the system? I’m not sure. Let me know. I’m sure that international finance people will be able to improve this plan, but we need a plan to minimize the chaos that a breakup of the euro will otherwise entail. I think this would help.

 

Written by Liam C Malloy

May 10, 2012 at 9:47 am

Posted in Eurozone

Is “Inflation…always and everywhere a monetary phenomenon”?

leave a comment »

Milton Friedman famously said that:

Inflation is always and everywhere a monetary phenomenon…

But that doesn’t seem to be the case recently:

The blue line is the annual percent change in M1, the green line is the annual percent change in M2, and the red line is the annual percent change in the CPI, or inflation. Inflation has remained muted while the money supply has increased dramatically over the last 4 years. The correlation between the percent change in M1 and the CPI is -0.48 and between M2 and the CPI is -0.19. Nonetheless, monetarists tell us to worry about inflation.

But is this really what Friedman meant? I don’t know, but Wikiquote gives us a more complete quote than I’ve given above:

Inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.

So what he seems to be saying is what seems to be the obvious truth that you can’t have inflation without an increase in the money supply. What he isn’t saying, at least in this quote, is that you will always get inflation with a large increase in the money supply. After all, if the money is not spent, then the price level won’t go up. And that seems to be the case recently. People and (especially) firms are hoarding money, mainly in the form of checking accounts, but also a bit in the form of currency. When they start spending again, the Fed is going to have to do some quick work to avoid inflation, but it certainly won’t be impossible.

Maybe I’m missing something obvious. After all, I’m not a monetary economist. But it seems to be like inflation is mainly caused by higher aggregate demand (along with things like indexation). It can be controlled by tightening monetary policy, but it can’t be produced by simply increasing the money supply.

That’s why I don’t really get the whole NGDP targeting argument. The argument, as I understand it, says that when real GDP growth is lower than the Fed would like, it should accept higher inflation. But low AD leads to lower inflation. How is the Fed supposed to get to 5% inflation in today’s economy? The money supply (M1) has increased by over 50% since 2007 and inflation has stayed well below 5%! What is the Fed supposed to do?

Really, I’m asking.

Written by Liam C Malloy

April 19, 2012 at 1:13 pm

Posted in Uncategorized

Republican “Intellectuals” View Life as a Game

leave a comment »

Alec MacGillis writes about Mankiw’s defense of the tax treatment of carried interest. At the end he (doesn’t) link to Matt O’Brien who links to a Mankiw blog post saying the exact opposite.

This is what I mean when I say that those academics and other public intellectuals on the right view this whole thing as a game. It’s not about doing what’s right and it’s not even about doing what you believe to be true. It’s about confusing the issue so that your masters can do what they want.

But it’s not a game. Allowing hedge fund managers to pay 15% in federal income taxes instead of 35% robs the Treasury of much needed revenue, putting a burden on current and future tax payers. It also increases inequality in a way that simply didn’t happen fifty years ago.

Written by Liam C Malloy

March 6, 2012 at 1:01 pm

The Needed Assumptions of DSGE

leave a comment »

Noah Smith has a great post on whether or not macroeconomic models are biased in one political direction or the other. He reaches the (correct, in my opinion) conclusion that they do have a conservative bias, mainly because they are so simplified that there can be no useful role for government.

I think the most useful post is the list of assumptions in the original Kydland and Prescott Real Business Cycle paper. With so many assumptions, that model is basically useless. Smith goes on to point out that if you relaxed even the ones that are most problematic (in my opinion: representative agent modeling, rational expectations, and flexible prices) you quickly get a model that is completely unwieldy.

Why so use these models at all? They are mathematically complex, they are micro-founded (sort of), and they give little to no role for government.

Personally, I didn’t like these models when I learned them, and I still don’t find them useful. Give me a partial-equilibrium model with a little more realism any day.

Written by Liam C Malloy

February 21, 2012 at 10:37 am

Posted in Uncategorized

Libertarians and Feminism

with 2 comments

I just came across two different posts that show libertarians uncomfortable stance toward feminism. The first was from Noah Smith showing that Peter Thiel thinks that allowing women to vote was a bad thing (at least for libertarians).

The second was from Mike Konczal that showed how uncomfortable women’s rights and female sexuality made Ludwig von Mises.

From a purely philosophical point of view, you would think that libertarians would welcome feminism as it increased the amount of freedom in the world by opening up new opportunities and giving new rights to half the population.

But libertarians (often, not always) seem to worry about protecting the rights and liberties of themselves and people like them, rather than extending them to others. Thus we end up with misogynistic and racist white male libertarians. One question is how women and minorities can feel comfortable in these groups.

 

Written by Liam C Malloy

February 13, 2012 at 4:35 pm

Posted in Libertarians

Follow

Get every new post delivered to your Inbox.