A second new paper: "A fiscal theory of monetary policy with partially repaid long-term debt."
By "fiscal theory of monetary policy" I mean a model with standard DSGE ingredients, including inertemporal optimization and market clearing, monetary policy described by interest rate targets, price or other frictions, but closed by fiscal theory, "active" fiscal policy rather than "active" monetary policy.
I aim to build a standard simple but somewhat realistic model of this sort, a parallel to the three equation textbook model that has been part of the new-Keynesian tool kit since the 1990s. I keep the model as simple and standard as possible, so the effect of the innovations one the fiscal side are clearer.
Two parts of the specification are central. First, long-term debt allows the model to produce a negative response of inflation to interest rates. Long-term debt also allows a fiscal shock to result in a protracted inflation, which slowly devalues long term bonds, rather than a price level jump.
Second, and most important, the paper writes down a process for fiscal surpluses in which today's deficits are partially repaid by tomorrow's surpluses. Look quickly at the surplus response functions in my last post. When the government runs a deficit, it reliably runs subsequent surpluses that partially repay some of the accumulated debt. The surplus is not an AR(1)! It has an s-shaped response function.
So if you want a realistic fiscal theory model, you need a surplus with an s-shaped response function, but you need to keep "active" fiscal policy. This combination is the central innovation of the paper.
Showing posts with label Academic Articles. Show all posts
Showing posts with label Academic Articles. Show all posts
Friday, February 7, 2020
Wednesday, February 5, 2020
New Paper -- the fiscal roots of inflation
I recently finished drafts of a few academic papers that blog readers might find interesting. Today, "The Fiscal Roots of Inflation."
The government debt valuation equation says that the real value of nominal debt equals the present value of surpluses. So, when there is inflation, the real value of nominal debt declines. Does that decline come about by lower future surpluses, or by a higher discount rate? You can guess the answer -- a higher discount rate.
Though to me this is interesting for how to construct fiscal theory models in which changes in the present value of government debt cause inflation, the valuation equation is every bit as much a part of standard new-Keynesian models. So the paper does not take a stand on causality.
Here is an example of the sort of puzzle the paper addresses. Think about 2008. There was a big recession. Deficits zoomed, through bailout, stabilizers, and deliberate stimulus. Yet inflation.. declined. So how does the government debt valuation equation work? Well, maybe today's deficits are bad, but they came with news of better future surpluses. That's hard to stomach. And it isn't true in the data. Well, real interest rates declined and sharply. The discount rate for government debt declined, which raises the value of government debt, even if expected future surpluses are unchanged or declined. With a lower discount rate, government debt is more valuable. If the price level does not change, people want to buy less stuff and more government debt. That's lower aggregate demand, which pushes the price level down. Does this story bear out, quantitatively, in the data? Yes.
If you don't like discount rates and forward looking behavior, you can put the same observation in ex-post terms. When there is a big deficit, the value of debt rises. How, on average, does the debt-GDP ratio come back down on average? Well, the government could run big surpluses -- raise taxes, cut spending to pay off the debt. That turns out not to be the case. There could be a surge of economic growth. Maybe the stimuluses and infrastructure spending all pay off. That turns out not to be the case. Or, the real rate of return on government bonds could go down, so that debt grows at a lower rate. That turns out to be, on average and therefore predictably, the answer.
Identities
OK, to work. The paper starts by developing a Cambpell-Shiller type identity for government debt. This works also for arbitrary maturity structures of the debt. Corresponding to the Campbell-Shiller return linearization, $$ \rho v_{t+1}=v_{t}+r_{t+1}^{n}-\pi_{t+1}-g_{t+1}-s_{t+1}. $$ The log debt to GDP ratio at the end of period \(t+1\), \(v_{t+1}\), is equal to its value at the end of period \(t\), \(v_{t}\), increased by the log nominal return on the portfolio of government bonds \(r_{t+1}^{n}\) less inflation \(\pi_{t+1}\), less log GDP growth \(g_{t+1}\), and less the real primary surplus to GDP ratio \(s_{t+1}\). Surpluses, unlike dividends, can be negative, so I don't take the log here. This surplus is scaled to have units of surplus to value, so a 1% change in "surplus" changes the log value of debt by 1%. I use this equation to measure the surplus.
Iterating forward, and imposing the transversality condition, we have a Campbell-Shiller style present value identity, $$ v_{t}=\sum_{j=1}^{\infty}\rho^{j-1}s_{t+j}+\sum_{j=1}^{\infty}\rho^{j-1}g_{t+j} -\sum_{j=1}^{\infty}\rho^{j-1}\left( r_{t+j}^{n}-\pi _{t+j}\right). $$ Take innovations \( \Delta E_{t+1} \equiv E_{t+1}-E_t \) and we have $$ \Delta E_{t+1}\pi_{t+1}-\Delta E_{t+1} r_{t+1}^{n}= -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1}s_{t+1+j} -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1} g_{t+1+j}+\sum_{j=1}^{\infty} \rho^{j} \Delta E_{t+1}\left( r_{t+1+j}^{n}-\pi_{t+1+j}\right) $$ Unexpected inflation devalues bonds. So it must come with a decline in surpluses, a rise in the discount rate, or a decline in bond prices. Notice the value of debt disappeared, which is handy.
The bond return comes from future expected returns or inflation, so it's nice to get rid of that too. With a geometric maturity structure in which the face value of bonds of \(j\) maturity is \(\omega^j\), a high bond return today must come from lower bond returns in the future. $$ \Delta E_{t+1}r_{t+1}^{n} = -\sum_{j=1}^{\infty}\omega^{j}\Delta E_{t+1} r_{t+1+j}^{n} =-\sum_{j=1}^{\infty}\omega^{j}\Delta E_{t+1}\left[ (r_{t+1+j}^{n}-\pi_{t+1+j})+\pi_{t+1+j}\right] $$ Substitute and we have the last and best identity $$ \sum_{j=0}^{\infty}\omega^{j} \Delta E_{t+1}\pi_{t+1+j} = -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1}s_{t+1+j} -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1}g_{t+1+j} +\sum_{j=1}^{\infty} (\rho^{j} -\omega^{j})\Delta E_{t+1}\left( r_{t+1+j}^{n}-\pi_{t+1+j}\right) . $$ With long-term debt a weighted sum of current and future inflation corresponds to changes in expected surpluses and discount rates. A fiscal shock can result in future inflation, thereby falling on today's long term bonds. Equivalently, a surprise deficit today \(s_{t+1}\) must be met by future surpluses, by lower returns, or by devaluing outstanding bonds, so that the debt/GDP ratio is reestablished.
Results
I ran a VAR and computed the responses to various shocks.
Here is the response to an inflation shock - -an unexpected movement \(\Delta E_1 \pi_1\). All other variables may move at the same time as the inflation shock.
Inflation is persistent, so a 1% inflation shock is about a 1.5% cumulative inflation shock, weighted
by the maturity of outstanding debt.
So, where is the 1.5% decline in present value of surpluses? Which terms of the identity matter?
Inflation does come with persistent deficits here. The sample is 1947-2018, so a lot of the inflation shocks come in the 1970s. You might raise three cheers for the fiscal theory, but not so fast. The deficits turn around and become surpluses. The sum of all surpluses term in the identity is a trivial -0.06, effectively zero. These deficits are essentially all paid back by subsequent surpluses.
Growth declines by half a percentage point cumulatively, accounting for 2/3 of the inflation. And the discount rate rises persistently. Two thirds of the devaluation of debt that inflation represents comes from higher real expected returns on government bonds, which in turn means higher interest rates that don't match inflation. (More graphs in the paper.)
Growth here is negatively correlated with inflation, which is true of the overall sample, but not of the story I started out with. What happens in a normal recession, that features lower inflation and lower output? Let's call it an aggregate demand shock. To measure such an event, I simply defined a shock that moves both output and inflation down by 1%. Here are the responses to this "recession shock."
Inflation and output go down now, by 1%, and by construction. That's how I defined the shock. This is a recession with low growth, low inflation, and deficits. Not shown, interest rates all decline too.
So where does the low inflation come from in the above decomposition. Do today's deficits signal future surpluses? Yes, a bit. But not enough -- the cumulative sum of surpluses is -1.15% On its own, deficits should cause 1% inflation, the fiscal theory puzzle that started me out in this whole business. Growth quickly recovers, but is not positive for a sustained period. Like 2008, we see a basically downward shift in the level of GDP. That contributes another 1% inflationary force. The discount rate falls however, so strongly as to raise the real value debt by almost 5 percentage points! That overcomes the inflationary forces and accounts for the deflation.
Here is a plot of the interest rates in response to the same shock. i is the three month rate, y is the 10 year rate, and rn is the return on the government bond portfolio. Yes, interest rates at all maturities jump down in this recession. Sharply lower rates mean a one-period windfall for the owners of long term bonds, then expected bond returns fall too.
The point
Discount rates matter. If you want to understand the fiscal foundations of inflation, you have to understand the government debt valuation equation. Inflation and deflation over the cycle is not driven by changing expected surpluses. If you want to view it "passively," inflation and deflation over the cycle does not result in passive policy accommodation through taxes, as most footnotes presume. The fiscal roots (or consequences) of inflation over the cycle are the strong variation in discount rates -- expected returns.
The fiscal process
Notice that the response of primary surpluses in all these graphs is s-shaped. Primary surpluses do not follow an AR(1) type process. In response to today's deficits, there is eventually a shift to a long string of surpluses that partially repay much though not all of that debt. This seems completely normal, except that so many models specify AR(1) style processes for fiscal surpluses. Surely that is a huge mistake. Stay tuned. The next paper shows how to put an s-shaped surplus process in a model and why it is so important to do so.
Comments on the paper are most welcome.
The government debt valuation equation says that the real value of nominal debt equals the present value of surpluses. So, when there is inflation, the real value of nominal debt declines. Does that decline come about by lower future surpluses, or by a higher discount rate? You can guess the answer -- a higher discount rate.
Though to me this is interesting for how to construct fiscal theory models in which changes in the present value of government debt cause inflation, the valuation equation is every bit as much a part of standard new-Keynesian models. So the paper does not take a stand on causality.
Here is an example of the sort of puzzle the paper addresses. Think about 2008. There was a big recession. Deficits zoomed, through bailout, stabilizers, and deliberate stimulus. Yet inflation.. declined. So how does the government debt valuation equation work? Well, maybe today's deficits are bad, but they came with news of better future surpluses. That's hard to stomach. And it isn't true in the data. Well, real interest rates declined and sharply. The discount rate for government debt declined, which raises the value of government debt, even if expected future surpluses are unchanged or declined. With a lower discount rate, government debt is more valuable. If the price level does not change, people want to buy less stuff and more government debt. That's lower aggregate demand, which pushes the price level down. Does this story bear out, quantitatively, in the data? Yes.
If you don't like discount rates and forward looking behavior, you can put the same observation in ex-post terms. When there is a big deficit, the value of debt rises. How, on average, does the debt-GDP ratio come back down on average? Well, the government could run big surpluses -- raise taxes, cut spending to pay off the debt. That turns out not to be the case. There could be a surge of economic growth. Maybe the stimuluses and infrastructure spending all pay off. That turns out not to be the case. Or, the real rate of return on government bonds could go down, so that debt grows at a lower rate. That turns out to be, on average and therefore predictably, the answer.
Identities
OK, to work. The paper starts by developing a Cambpell-Shiller type identity for government debt. This works also for arbitrary maturity structures of the debt. Corresponding to the Campbell-Shiller return linearization, $$ \rho v_{t+1}=v_{t}+r_{t+1}^{n}-\pi_{t+1}-g_{t+1}-s_{t+1}. $$ The log debt to GDP ratio at the end of period \(t+1\), \(v_{t+1}\), is equal to its value at the end of period \(t\), \(v_{t}\), increased by the log nominal return on the portfolio of government bonds \(r_{t+1}^{n}\) less inflation \(\pi_{t+1}\), less log GDP growth \(g_{t+1}\), and less the real primary surplus to GDP ratio \(s_{t+1}\). Surpluses, unlike dividends, can be negative, so I don't take the log here. This surplus is scaled to have units of surplus to value, so a 1% change in "surplus" changes the log value of debt by 1%. I use this equation to measure the surplus.
Iterating forward, and imposing the transversality condition, we have a Campbell-Shiller style present value identity, $$ v_{t}=\sum_{j=1}^{\infty}\rho^{j-1}s_{t+j}+\sum_{j=1}^{\infty}\rho^{j-1}g_{t+j} -\sum_{j=1}^{\infty}\rho^{j-1}\left( r_{t+j}^{n}-\pi _{t+j}\right). $$ Take innovations \( \Delta E_{t+1} \equiv E_{t+1}-E_t \) and we have $$ \Delta E_{t+1}\pi_{t+1}-\Delta E_{t+1} r_{t+1}^{n}= -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1}s_{t+1+j} -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1} g_{t+1+j}+\sum_{j=1}^{\infty} \rho^{j} \Delta E_{t+1}\left( r_{t+1+j}^{n}-\pi_{t+1+j}\right) $$ Unexpected inflation devalues bonds. So it must come with a decline in surpluses, a rise in the discount rate, or a decline in bond prices. Notice the value of debt disappeared, which is handy.
The bond return comes from future expected returns or inflation, so it's nice to get rid of that too. With a geometric maturity structure in which the face value of bonds of \(j\) maturity is \(\omega^j\), a high bond return today must come from lower bond returns in the future. $$ \Delta E_{t+1}r_{t+1}^{n} = -\sum_{j=1}^{\infty}\omega^{j}\Delta E_{t+1} r_{t+1+j}^{n} =-\sum_{j=1}^{\infty}\omega^{j}\Delta E_{t+1}\left[ (r_{t+1+j}^{n}-\pi_{t+1+j})+\pi_{t+1+j}\right] $$ Substitute and we have the last and best identity $$ \sum_{j=0}^{\infty}\omega^{j} \Delta E_{t+1}\pi_{t+1+j} = -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1}s_{t+1+j} -\sum_{j=0}^{\infty} \rho^{j} \Delta E_{t+1}g_{t+1+j} +\sum_{j=1}^{\infty} (\rho^{j} -\omega^{j})\Delta E_{t+1}\left( r_{t+1+j}^{n}-\pi_{t+1+j}\right) . $$ With long-term debt a weighted sum of current and future inflation corresponds to changes in expected surpluses and discount rates. A fiscal shock can result in future inflation, thereby falling on today's long term bonds. Equivalently, a surprise deficit today \(s_{t+1}\) must be met by future surpluses, by lower returns, or by devaluing outstanding bonds, so that the debt/GDP ratio is reestablished.
Results
I ran a VAR and computed the responses to various shocks.
Here is the response to an inflation shock - -an unexpected movement \(\Delta E_1 \pi_1\). All other variables may move at the same time as the inflation shock.
Inflation is persistent, so a 1% inflation shock is about a 1.5% cumulative inflation shock, weighted
by the maturity of outstanding debt.
So, where is the 1.5% decline in present value of surpluses? Which terms of the identity matter?
Inflation does come with persistent deficits here. The sample is 1947-2018, so a lot of the inflation shocks come in the 1970s. You might raise three cheers for the fiscal theory, but not so fast. The deficits turn around and become surpluses. The sum of all surpluses term in the identity is a trivial -0.06, effectively zero. These deficits are essentially all paid back by subsequent surpluses.
Growth declines by half a percentage point cumulatively, accounting for 2/3 of the inflation. And the discount rate rises persistently. Two thirds of the devaluation of debt that inflation represents comes from higher real expected returns on government bonds, which in turn means higher interest rates that don't match inflation. (More graphs in the paper.)
Growth here is negatively correlated with inflation, which is true of the overall sample, but not of the story I started out with. What happens in a normal recession, that features lower inflation and lower output? Let's call it an aggregate demand shock. To measure such an event, I simply defined a shock that moves both output and inflation down by 1%. Here are the responses to this "recession shock."
So where does the low inflation come from in the above decomposition. Do today's deficits signal future surpluses? Yes, a bit. But not enough -- the cumulative sum of surpluses is -1.15% On its own, deficits should cause 1% inflation, the fiscal theory puzzle that started me out in this whole business. Growth quickly recovers, but is not positive for a sustained period. Like 2008, we see a basically downward shift in the level of GDP. That contributes another 1% inflationary force. The discount rate falls however, so strongly as to raise the real value debt by almost 5 percentage points! That overcomes the inflationary forces and accounts for the deflation.
Here is a plot of the interest rates in response to the same shock. i is the three month rate, y is the 10 year rate, and rn is the return on the government bond portfolio. Yes, interest rates at all maturities jump down in this recession. Sharply lower rates mean a one-period windfall for the owners of long term bonds, then expected bond returns fall too.
Discount rates matter. If you want to understand the fiscal foundations of inflation, you have to understand the government debt valuation equation. Inflation and deflation over the cycle is not driven by changing expected surpluses. If you want to view it "passively," inflation and deflation over the cycle does not result in passive policy accommodation through taxes, as most footnotes presume. The fiscal roots (or consequences) of inflation over the cycle are the strong variation in discount rates -- expected returns.
The fiscal process
Notice that the response of primary surpluses in all these graphs is s-shaped. Primary surpluses do not follow an AR(1) type process. In response to today's deficits, there is eventually a shift to a long string of surpluses that partially repay much though not all of that debt. This seems completely normal, except that so many models specify AR(1) style processes for fiscal surpluses. Surely that is a huge mistake. Stay tuned. The next paper shows how to put an s-shaped surplus process in a model and why it is so important to do so.
Comments on the paper are most welcome.
Tesla Bubble?
Paul Vigna in the Wall Street Journal
Holman Jenkins:
This pattern happens over and over (and over and over) again in financial markets. Surely we can do better as an "explanation" than "people are dumb." Or, as Lamont and Thaler put it so nicely,
Stocks as money points out these events do not happen in isolation. High prices are only one symptom. They always occur with 1) huge price volatility (check) 2) huge share turnover (check) 3) impediments to short-selling, especially at a longer horizon (check, more below) 4) in an asset where there is a lot of disagreement, a lot of potential news, a lot of different opinions about long run value. Bubbles do not happen in regulated public utility stocks.
"People are dumb and will pay too much for flashy stuff" does not explain why they should, a week later, change their minds and sell. It does not explain why high prices only happen with the other four.
So why would anyone buy Tesla?
 |
| Source: Wall Street Journal |
"Tesla TSLA -18.51% Inc.’s shares rose 14% Tuesday to $887.06. They have surged 56% in the past week and have nearly quadrupled since early October. Those outsize gains don’t match Tesla’s more modest fundamentals, which include annual losses."
"They do, however, resemble any number of other assets that have experienced prolonged bubbles, including shares of Qualcomm Inc. and other tech stocks of the dot-com era; oil in 2008 and bitcoin in 2017."At these prices, Tesla is worth more than Ford and GM combined.
Holman Jenkins:
Tesla can earn a lot more profit per car, and can sell a lot more cars than it does now, and still its stock is priced as if its future profits will be coming from some unnamed something that is not the car-making business.A correspondent:
I just looked at the minute by minute data for TSLA. In the last 12 minutes of trading volume was 4.5 million shares, high price was 967, low price (in the last 12 minutes not the day) came 5 minutes later at 860, closed at 887.06.
Shares outstanding is 180 million so the move from 969 to 860 erased 20 billion of market cap, in five minutes, on no news. These numbers are so crazy they seem almost meaningless but they make for a good sound byte in a video.And implicitly (he's more polite)
So Mr. Efficient Market, what do you make of that?I can't resist the temptation to plug an old paper, that I have long wanted to return to, "Stocks as Money." I wrote it in response to the internet boom and bust, and the excellent Owen Lamont -Richard Thaler "Can the market add and subtract" in particular.
This pattern happens over and over (and over and over) again in financial markets. Surely we can do better as an "explanation" than "people are dumb." Or, as Lamont and Thaler put it so nicely,
one needs investors who are (in our specific case) irrational, woefully uninformed, endowed with strange preferences, or for some other reason willing to hold over-priced assets.Patterns that are repeated over and over again need ether irredeemable human folly -- not much of an "explanation" as it can explain anything -- or economics, a model by which rules of the game produce a strange outcome despite people in the game understanding the game and where it ends. That's what "stocks as money" suggested.
Stocks as money points out these events do not happen in isolation. High prices are only one symptom. They always occur with 1) huge price volatility (check) 2) huge share turnover (check) 3) impediments to short-selling, especially at a longer horizon (check, more below) 4) in an asset where there is a lot of disagreement, a lot of potential news, a lot of different opinions about long run value. Bubbles do not happen in regulated public utility stocks.
"People are dumb and will pay too much for flashy stuff" does not explain why they should, a week later, change their minds and sell. It does not explain why high prices only happen with the other four.
So why would anyone buy Tesla?
Saturday, January 11, 2020
Wealth and taxes -- Overview
I thought that "wealth and taxes" would be a short blog post. It turned in to a 5 part series. Here's an overview, or table of contents in case the whole thing looks a bit indimidating. The most important one, really I think is Part V, "it's all political." The others build bit by bit, well, this can't be the answer and that can't be the answer, so what is the answer, and Part V finds it.
In Part I we met the fact that "wealth" is measured as "capitalized income," Y/r. But only some kinds of income Y and with discount rate choices r that blew up measured wealth inequality. I review the Smith, Zidar and Zwick paper that finds huge overstatements of inequality because wealthy people have a higher r than you and me.
In Part II we learned that a big reason wealth inequality widened is that interest rates fell and asset prices rose. If r falls, Y/r rises, but it's the same Y.
In Part III we noted the distinction between consumption, income and wealth inequality. Wealth is beyond badly measured as a measure of lifestyle. The computations ignore taxes and transfers, wildly blowing up measured inequality and rendering it a "problem" that ipso facto cannot be solved. Why concern ourselves with pre-tax wealth inequality, especially given that most wealth is reinvested in businesses that produce things and employ people?
In Part IV, we met the wealth tax. If the question is, how do we raise revenue for the government, either to spend or to transfer it, the wealth tax is a terrible idea, as it distorts the economy and leads to an evasion industry. A consumption tax is a much better idea.
In Part V I read Saez and Zucman's opeds, which finally tell us what the question is to which the wealth tax is the answer. Saez and Zucman want to confiscate billionaires' wealth, because they think billionaires have too much political power, billionaires all got their money unjustly, and somehow though big government cronyism is the problem, bigger government is the answer. The wealth tax is not designed to raise revenue -- it succeeds if it raises no revenue (after perhaps a one-time wealth grab) because the wealth it taxes has vanished. Well, at least it is a consistent view, decide if you buy the premises.
Update
CATO put out a much-improved version of this series. Html here and pdf here.
In Part I we met the fact that "wealth" is measured as "capitalized income," Y/r. But only some kinds of income Y and with discount rate choices r that blew up measured wealth inequality. I review the Smith, Zidar and Zwick paper that finds huge overstatements of inequality because wealthy people have a higher r than you and me.
In Part II we learned that a big reason wealth inequality widened is that interest rates fell and asset prices rose. If r falls, Y/r rises, but it's the same Y.
In Part III we noted the distinction between consumption, income and wealth inequality. Wealth is beyond badly measured as a measure of lifestyle. The computations ignore taxes and transfers, wildly blowing up measured inequality and rendering it a "problem" that ipso facto cannot be solved. Why concern ourselves with pre-tax wealth inequality, especially given that most wealth is reinvested in businesses that produce things and employ people?
In Part IV, we met the wealth tax. If the question is, how do we raise revenue for the government, either to spend or to transfer it, the wealth tax is a terrible idea, as it distorts the economy and leads to an evasion industry. A consumption tax is a much better idea.
In Part V I read Saez and Zucman's opeds, which finally tell us what the question is to which the wealth tax is the answer. Saez and Zucman want to confiscate billionaires' wealth, because they think billionaires have too much political power, billionaires all got their money unjustly, and somehow though big government cronyism is the problem, bigger government is the answer. The wealth tax is not designed to raise revenue -- it succeeds if it raises no revenue (after perhaps a one-time wealth grab) because the wealth it taxes has vanished. Well, at least it is a consistent view, decide if you buy the premises.
Update
CATO put out a much-improved version of this series. Html here and pdf here.
Thursday, January 9, 2020
Wealth and Taxes, Part IV
(This is Part IV of a series. Part III, and Part V. which has the punchline. See the overview for a summary.)
The Wealth Tax.
So, if arguing about the ill-defined and ill-measured distribution of wealth lies in service of the wealth tax, what is the question to which the wealth tax is an answer?
Revenue and Redistribution -- good and bad taxes
Preamble: Economists have no real professional expertise to object to redistribution, or argue for it. Swallow hard, you may not like it for political, moral or other reasons -- or you may be all for it for those reasons -- but admit economists economists have no special insights to the right amount of redistribution. Economics has one analysis to offer the world: incentives. (OK, and equilibrium.) If it were possible to take money from A and give it to B without creating any adverse incentives, we have no special standing to cheer or to object. Economics can tell us something about tax rates, but not much about taxes.
Thus the theory of optimal taxation is straightforward: how can the government raise a given tax revenue while generating the least perverse disincentives? The theory of optimal redistribution offers an additional wrinkle: how can the government give money away while generating the least perverse disincentives to recipients as well as payers?
The Wealth Tax.
So, if arguing about the ill-defined and ill-measured distribution of wealth lies in service of the wealth tax, what is the question to which the wealth tax is an answer?
Revenue and Redistribution -- good and bad taxes
Preamble: Economists have no real professional expertise to object to redistribution, or argue for it. Swallow hard, you may not like it for political, moral or other reasons -- or you may be all for it for those reasons -- but admit economists economists have no special insights to the right amount of redistribution. Economics has one analysis to offer the world: incentives. (OK, and equilibrium.) If it were possible to take money from A and give it to B without creating any adverse incentives, we have no special standing to cheer or to object. Economics can tell us something about tax rates, but not much about taxes.
Thus the theory of optimal taxation is straightforward: how can the government raise a given tax revenue while generating the least perverse disincentives? The theory of optimal redistribution offers an additional wrinkle: how can the government give money away while generating the least perverse disincentives to recipients as well as payers?
Sunday, January 5, 2020
Conference announcement
If you're working on fiscal issues, especially fiscal theory of the price level, here is a good conference you should submit to or attend. Don't wait, the deadline is today. Yes, this is self-interested -- I'm going and giving a talk so it's entirely in my interest that the other papers are interesting! The conference website is here
Thursday, January 2, 2020
Wealth and taxes, part I
(This is Part I of a series. See the overview for a summary. The punchline comes in Part V.)
Last November I had the pleasure of discussing "Top Wealth in the United States: New Estimates and Implications for Taxing the Rich" a very nice paper by Matthew Smith, Owen Zidar and Eric Zwick at the NBER asset pricing meetings, presented by Eric. The paper prompts a series of blog posts on wealth distribution and wealth taxes. I'll try to stick to points that haven't been made a hundred times already.
The paper mostly examines Saez and Zucman's 2016 QJE paper on wealth inequality. As many others have found, the Saez Zucman numbers are, ... let's say somewhat overstated.
Their bottom line is to cut Saez and Zucman in half. As I read the paper I think this is conservative -- and when we ask the obvious questions that the whole enterprise begs to be asked (which Smith et al don't do, but I will) a chasm of emptiness opens up, and the questions end up emptier than their answers.
The first thing you have to understand is the nature of wealth. Here is most people's impression of what wealth is:
That's not it at all. As Zwick et al say,
You may have wondered, if we're just going to mulitply income by a number and call it wealth, why are we bothering to measure the wealth distribution at all? Let's just use the income distribution! You get one answer here -- if you call it wealth you get to multiply by 113! Since only some kinds of income get this treatment, kinds that are more likely to be held by wealthy people, that makes the numbers look much more unequal.
Smith et al's point though is not this basic one. Rather they look carefully at the calculation. This calculation assumes that all "fixed income" assets pay the same, low, rate of interest. Another well established fact is that rich people get better rates of return on their assets.
Here is Smith et al's plot of the actual rate of return that people earn on their fixed income investments. The uber wealthy earn 6% on their fixed income investments. This is not a small effect. In our capitalization factors, wealth = income / discount rate,
Changing from a 0.01 discount rate to a 0.06 discount rate pulls the wealth estimate per dollar of income down from 100 to 16.7. That's a lot. Smith et al:
(Note the irony. People who worry about wealth inequality are usually bemoan the fact that rich people earn higher returns on average than not so rich people, as it apparently will make inequality worse over time. But the same higher average return must mean a lower multiples for converting income to wealth. You just can't have it both ways.
Higher returns are not some evil plot. The largely come from the fact that rich people buy riskier assets, like stocks and junk bonds, and less rich people buy safer but lower yielding assets like bank accounts. OK, It is to some extent a plot. Lots of regulations prohibit lower income people from buying the kinds of assets that make rich people richer in the name of consumer protection. The SEC is loosening some of these regulations.)
Beyond fixed income, the capitalization game gets even muddier, in both papers. What income flow are you going to capitalize?
The bottom line? The game, as announced by Saez and Zucman is this: We start with the pretax value of “capital” income, including asset income, proprietor income and partnership income, but not labor income (wages, bonuses, etc) or social security income. We multiply by various huge 1/r numbers to call them "wealth". By doing that and using low r numbers, the "wealth" distribution looks much more extreme than the income distribution. As you can see the 1/r assumption allows great latitude in how this calculation is going to come out.
****
I spent a lot of time in asset pricing, and this paper was presented at an asset pricing meeting, so let me offer a little bit of what asset pricing has to say about these kinds of procedures.
The real capitalization formula is
The function 1/(r-g) is very sensitive to r and g, especially for low discount rates like the 1% we were using for bonds. Going down from 2% to 1% doubles the value. So, if you want to fiddle with values, fiddle with discount rates.
The right discount rate is much higher for risky assets than risk free assets. Lots of people discount things with stock market risk using interest rates, and get absurdly too high values.
If you put the 20 best financial economists in the world together in a room, gave them all of a company's cash flow information, they could not come within a factor of 3 of the actual stock market value. "Valuation" mostly consists of fiddling with discount rates to get the "right" answer. Maybe "multiples" isn't so bad after all.
In short, capitalizing income to get any sense of "wealth" is an inherently... absurdly imprecise game.
***
I don't mean to sound critical of Smith et al. They're doing the best they can given the Zucman and Saez rules of the game. But a little peek into this sausage factory should leave you wondering, just why are these the rules of the game? Why do we care (should we care) so much about the distribution of something that is essentially impossible to measure or define? If you are making money was a partner in an LLC you help to run, why should anyone care about a fictitious accounting "value" of that partnership? You can't sell it!
Why start with pretax income? If you pay half your income in taxes, does that not halve the value of the asset? Why does "wealth" include the value of proprietor and partnership income but not labor income or social security income?
These are good questions for the next few blog posts. Stay tuned.
On to Part II
Last November I had the pleasure of discussing "Top Wealth in the United States: New Estimates and Implications for Taxing the Rich" a very nice paper by Matthew Smith, Owen Zidar and Eric Zwick at the NBER asset pricing meetings, presented by Eric. The paper prompts a series of blog posts on wealth distribution and wealth taxes. I'll try to stick to points that haven't been made a hundred times already.
The paper mostly examines Saez and Zucman's 2016 QJE paper on wealth inequality. As many others have found, the Saez Zucman numbers are, ... let's say somewhat overstated.
Their bottom line is to cut Saez and Zucman in half. As I read the paper I think this is conservative -- and when we ask the obvious questions that the whole enterprise begs to be asked (which Smith et al don't do, but I will) a chasm of emptiness opens up, and the questions end up emptier than their answers.
The first thing you have to understand is the nature of wealth. Here is most people's impression of what wealth is:
That's not it at all. As Zwick et al say,
“Less than half of top wealth takes the form of liquid securities with clear market values”
So, the question is how do we measure the "wealth" that is not liquid securities with clear market values, like the profits of privately owned businesses? And, given that there is not US data on wealth (yet, thank goodness), even the part that is a security is hard to measure.
Enter "capitalization." The main idea in Saez and Zucman, reexamined by Smith et al., is that we measure "wealth" by measuring income, and then translating that income to wealth by assuming it will last forever and discounting it at some rate. In equations
Wealth = Income / discount rate
We have data from the IRS on income. So, let's follow along on Zwick et al.'s best story, how we find wealth invested in bonds from IRS individual interest income data and total bonds outstanding data:
“In 2014, the aggregate flow of [taxable] interest income was $98B, and the stock of fixed income wealth was $11T. The ratio gives the average yield, r = $98B/$11T = 0.89%. Using this yield to capitalize income amounts to multiplying every dollar of interest income by 1/0.89% = 113 to estimate fixed income wealth. … Implementing equation (4) for fixed income gives an estimate of top fixed income wealth of $42B × 113 = $4.7T of fixed income wealth held by the top 0.1%. The bottom 99.9% estimate is $56B × 113 = $6.4T .My emphasis.
You may have wondered, if we're just going to mulitply income by a number and call it wealth, why are we bothering to measure the wealth distribution at all? Let's just use the income distribution! You get one answer here -- if you call it wealth you get to multiply by 113! Since only some kinds of income get this treatment, kinds that are more likely to be held by wealthy people, that makes the numbers look much more unequal.
Smith et al's point though is not this basic one. Rather they look carefully at the calculation. This calculation assumes that all "fixed income" assets pay the same, low, rate of interest. Another well established fact is that rich people get better rates of return on their assets.
Here is Smith et al's plot of the actual rate of return that people earn on their fixed income investments. The uber wealthy earn 6% on their fixed income investments. This is not a small effect. In our capitalization factors, wealth = income / discount rate,
1/0.01 = 100
1/0.06 = 16.7
“the adjustment reduces the top capitalization factor—and thus estimated top fixed income wealth—by a factor of 4.7, or 80%”This is huge, to say the least.
(Note the irony. People who worry about wealth inequality are usually bemoan the fact that rich people earn higher returns on average than not so rich people, as it apparently will make inequality worse over time. But the same higher average return must mean a lower multiples for converting income to wealth. You just can't have it both ways.
Higher returns are not some evil plot. The largely come from the fact that rich people buy riskier assets, like stocks and junk bonds, and less rich people buy safer but lower yielding assets like bank accounts. OK, It is to some extent a plot. Lots of regulations prohibit lower income people from buying the kinds of assets that make rich people richer in the name of consumer protection. The SEC is loosening some of these regulations.)
Beyond fixed income, the capitalization game gets even muddier, in both papers. What income flow are you going to capitalize?
“In the case of C-corporation equities, the income flow is dividends plus [realized] capital gains."I think that's an accounting mistake, common in this literature. You cannot take the realized capital gains as an "income" flow for capitalization purposes. Suppose you buy a stock for $1, and it grows to $100. You sell $10 of the stock, but now you only have $90 left. You can't keep doing this forever, as the capitalization assumes. That's fundamentally different than the company is worth $100, makes a $10 profit and gives you a $10 dividend. I'll be curious to hear from better accountants than I whether you can sensibly capitalize realized capital gains. Onwards...
For S-corporation equities, the income flow is S-corporation income. For proprietor and partnership wealth, the income flow is the sum of proprietor income and partnership income [ “capital” income?]. In the case of real estate, property tax is capitalized to estimate housing assets ….”Ok, that's income, what is the discount rate?
“Private business returns are harder to estimate than fixed income returns because private business wealth is harder to observe than fixed income wealth…We focus on multiple-based valuation models”So we go from multiples to estimate a multiple... This all seems rather circular.
The bottom line? The game, as announced by Saez and Zucman is this: We start with the pretax value of “capital” income, including asset income, proprietor income and partnership income, but not labor income (wages, bonuses, etc) or social security income. We multiply by various huge 1/r numbers to call them "wealth". By doing that and using low r numbers, the "wealth" distribution looks much more extreme than the income distribution. As you can see the 1/r assumption allows great latitude in how this calculation is going to come out.
****
I spent a lot of time in asset pricing, and this paper was presented at an asset pricing meeting, so let me offer a little bit of what asset pricing has to say about these kinds of procedures.
The real capitalization formula is
P/D = 1/(r-g)
the price - dividend ratio is equal to one over the difference of the discount rate and the growth rate of dividends. Shhh! If the wealth inequality crowd realizes they can subtract g their multipliers will explode! (Joke. Of course we always use the right numbers)
The function 1/(r-g) is very sensitive to r and g, especially for low discount rates like the 1% we were using for bonds. Going down from 2% to 1% doubles the value. So, if you want to fiddle with values, fiddle with discount rates.
The right discount rate is much higher for risky assets than risk free assets. Lots of people discount things with stock market risk using interest rates, and get absurdly too high values.
If you put the 20 best financial economists in the world together in a room, gave them all of a company's cash flow information, they could not come within a factor of 3 of the actual stock market value. "Valuation" mostly consists of fiddling with discount rates to get the "right" answer. Maybe "multiples" isn't so bad after all.
In short, capitalizing income to get any sense of "wealth" is an inherently... absurdly imprecise game.
***
I don't mean to sound critical of Smith et al. They're doing the best they can given the Zucman and Saez rules of the game. But a little peek into this sausage factory should leave you wondering, just why are these the rules of the game? Why do we care (should we care) so much about the distribution of something that is essentially impossible to measure or define? If you are making money was a partner in an LLC you help to run, why should anyone care about a fictitious accounting "value" of that partnership? You can't sell it!
Why start with pretax income? If you pay half your income in taxes, does that not halve the value of the asset? Why does "wealth" include the value of proprietor and partnership income but not labor income or social security income?
These are good questions for the next few blog posts. Stay tuned.
On to Part II
Monday, November 25, 2019
Childbirth and crime
Family formation and crime is the title of a very nice new paper by Maxim Massenkoff and Evan K. Rose. (HT Alex Tabarrok at Marginal Revolution, which also has great commentary.)
The graphs speak for themselves. Go to the paper to look at them all. A few select ones:
Arrests fall by half, starting when mothers know they are pregnant:
(The paper presents more accurate but less interpretable event study coefficients. If you know what that means, go look at the paper.)
Father's crime drops too:
This decline isn't as steep. But first of all note it's all fathers, married or no, and second third and more kids. Then look at the huge difference in vertical scale. Women go from 3 to 2 economic offenses per 10,000. Men go from 20 to12 economic offenses per 10,000. This is a huge reduction in crime rates.
The graphs speak for themselves. Go to the paper to look at them all. A few select ones:
Arrests fall by half, starting when mothers know they are pregnant:
(The paper presents more accurate but less interpretable event study coefficients. If you know what that means, go look at the paper.)
Father's crime drops too:
This decline isn't as steep. But first of all note it's all fathers, married or no, and second third and more kids. Then look at the huge difference in vertical scale. Women go from 3 to 2 economic offenses per 10,000. Men go from 20 to12 economic offenses per 10,000. This is a huge reduction in crime rates.
Monday, July 22, 2019
Everything is f***d
The most hilarious course syllabus I've seen in a while, from Professor Sanjay Srivastava at the University of Oregon.
...
....
The point is serious, going well beyond the replication problem. Meta-analyses just repeat the same mistakes a hundred times.
"Office hours: Held on Twitter at your convenience." I love it.
...
The point is serious, going well beyond the replication problem. Meta-analyses just repeat the same mistakes a hundred times.
"Office hours: Held on Twitter at your convenience." I love it.
Saturday, July 20, 2019
New Papers
I've been remiss about blogging lately while I finished two new papers, "The Fiscal Roots of Inflation," and "The Value of Government Debt." I'm posting here for those who might be interested, and I appreciate comments.
Both papers apply asset pricing variance decompositions to questions of government finance and inflation. The inflation paper is part of the long-running fiscal theory of the price level project. (Note: this post uses MathJax which may not show properly on all devices.)
I start by deriving an analogue to the Campbell-Shiller linearization of the return identity:
\[ \beta v_{t+1}=v_{t}+r_{t+1}^{n}-\pi_{t+1}-g_{t+1}-s_{t+1}. \] The log debt to GDP ratio at the end of period \(t+1\), \(v_{t+1}\), is equal to its value at the end of period t, \(v_{t}\), increased by the log nominal return on the portfolio of government bonds \(r_{t+1}^{n}\) less inflation \(\pi_{t+1}\), less log GDP growth \(g_{t+1}\), and less the primary surplus \(s_{t+1}\). The "surplus" in this linearization is the surplus to GDP ratio, divided by the steady state debt to GDP ratio. It's not a log, so it can be negative. \(\beta = e^{-(r-g)}\) is a discount rate corresponding to the steady state real rate \(r\) less GDP growth rate \(g\).
Iterating forward, the present value identity is \begin{equation} v_{t}=\sum_{j=1}^{\infty}\beta^{j-1}\left[ s_{t+j}-\left( r_{t+j}^{n}-\pi_{t+j}+g_{t+j}\right) \right] .\label{pvsy}% \end{equation} I simplify by using \(\beta=1\) as the point of linearization. 1 vs. 0.99 doesn't make any significant difference for empirical purposes.
Now apply standard asset pricing ideas. To focus on inflation, in "Fiscal Roots" I take the time \(t+1\) innovation of the present value identity, \(\Delta E_{t+1}\equiv E_{t+1}-E_{t}\). Rearranging, we have the unexpected inflation identity, \begin{align} & \Delta E_{t+1}\pi_{t+1}-\Delta E_{t+1}\left( r_{t+1}^{n}-g_{t+1}\right) \label{Epiintro}\\ & =-\sum_{j=0}^{\infty}\Delta E_{t+1}s_{t+1+j}+\sum_{j=1}^{\infty}\Delta E_{t+1}\left( r_{t+1+j}^{n}-\pi_{t+1+j}-g_{t+1+j}\right) .\nonumber \end{align} A decline in the present value of surpluses, coming either from a change in expected surpluses or a rise in their discount rates, must result in a lower real value of the debt. This reduction can come about by unexpected inflation, or by a decline in nominal long-term bond prices. The value of debt drops out, which is handy and simplifies matters.
Both papers apply asset pricing variance decompositions to questions of government finance and inflation. The inflation paper is part of the long-running fiscal theory of the price level project. (Note: this post uses MathJax which may not show properly on all devices.)
I start by deriving an analogue to the Campbell-Shiller linearization of the return identity:
\[ \beta v_{t+1}=v_{t}+r_{t+1}^{n}-\pi_{t+1}-g_{t+1}-s_{t+1}. \] The log debt to GDP ratio at the end of period \(t+1\), \(v_{t+1}\), is equal to its value at the end of period t, \(v_{t}\), increased by the log nominal return on the portfolio of government bonds \(r_{t+1}^{n}\) less inflation \(\pi_{t+1}\), less log GDP growth \(g_{t+1}\), and less the primary surplus \(s_{t+1}\). The "surplus" in this linearization is the surplus to GDP ratio, divided by the steady state debt to GDP ratio. It's not a log, so it can be negative. \(\beta = e^{-(r-g)}\) is a discount rate corresponding to the steady state real rate \(r\) less GDP growth rate \(g\).
Iterating forward, the present value identity is \begin{equation} v_{t}=\sum_{j=1}^{\infty}\beta^{j-1}\left[ s_{t+j}-\left( r_{t+j}^{n}-\pi_{t+j}+g_{t+j}\right) \right] .\label{pvsy}% \end{equation} I simplify by using \(\beta=1\) as the point of linearization. 1 vs. 0.99 doesn't make any significant difference for empirical purposes.
Now apply standard asset pricing ideas. To focus on inflation, in "Fiscal Roots" I take the time \(t+1\) innovation of the present value identity, \(\Delta E_{t+1}\equiv E_{t+1}-E_{t}\). Rearranging, we have the unexpected inflation identity, \begin{align} & \Delta E_{t+1}\pi_{t+1}-\Delta E_{t+1}\left( r_{t+1}^{n}-g_{t+1}\right) \label{Epiintro}\\ & =-\sum_{j=0}^{\infty}\Delta E_{t+1}s_{t+1+j}+\sum_{j=1}^{\infty}\Delta E_{t+1}\left( r_{t+1+j}^{n}-\pi_{t+1+j}-g_{t+1+j}\right) .\nonumber \end{align} A decline in the present value of surpluses, coming either from a change in expected surpluses or a rise in their discount rates, must result in a lower real value of the debt. This reduction can come about by unexpected inflation, or by a decline in nominal long-term bond prices. The value of debt drops out, which is handy and simplifies matters.
Friday, June 7, 2019
Futures forecasts
Torsten Slok at DB updates this lovely graph on occasion. Here's what it means. Fed fund futures are essentially bets on where the Federal funds rate will be at various points in the future. Thus, you can read from the dashed lines the market's guess about where the federal funds rate will go -- assuming that the bets are priced to have an even chance of winning or losing.
Reading it that way, the market was systematically wrong from 2009 to 2016. It's something like springtime in Chicago -- this week, 40 degrees and raining. Next week, 75 and sunny. Week after week after week. In 2017, the market finally changed expectations to say, no, fed funds rates are not rising -- just in time to miss the actual rise in federal funds. Now, as in the blue line, market forecasts say there will be a big decline. But, as Torsten points out, why would the market be right today?
So what does this graph mean? Are market practitioners really that dumb? After all, there is a lot of money to be made here. When the graph is upward sloping -- as the entire yield curve was upward sloping from 2009-2016 -- and so long as rates don't rise, you can make a fortune borrowing short and lending long. And vice versa. In short, the difference between forward rate (right end of dashed lines) and spot rate (current fed funds rate) does a lousy job of forecasting where the spot rate will go -- and thus, mechanically, is a good signal of the extra return, positive or (lately) negative you will get by holding long-term bonds.
The pattern is actually widespread and longstanding. Starting in the late 70s and early 1980s, Gene Fama wrote a series of papers on it, short term bonds, money markets, foreign exchange, and (a favorite of mine) long term bonds (with Rob Bliss). Campbell and Shiller also found it in long term bonds, which Monika Piazzesi and I extended. Piazzesi and Swanson show the pattern in federal funds futures.
There are three potential stories:
One: the market is dumb. People are dumb. Well, that's a nice story that can "explain" just about anything. But if you're so smart why are you not rich. Behavioral finance isn't that empty, and searches for common patterns in dumbness. However this graph is the opposite of the usual behavioral claim, extrapolative expectations, excess belief in momentum. If there is a rejectable hypothesis in behavioral finance, this graph seems to reject it. (I welcome corrections to that statement in the comments.)
Two: there is a risk premium and it varies over time. For most of 2009-2015, the economy was depressed. People needed a good promised return, a coin more than 50/50, to hold the risk of long term bonds. Once we exit the recession, the opposite pattern holds. Long term bonds should pay less than short term bonds, and maybe now the yield curve is finally waking up to that pattern. Naturally, I'm attracted to this story, but I admit it's a bit strained late in the upslope period.
Three: exit and entry to recessions is something like a rare event, a Poisson process. Such a process is like computer failure. The chance of the event is always the same, and does not increase as the length of time goes by. Recovery could happen any time. In a second paper that's what Piazzesi and I seemed to find for this pattern in bond markets. The market forecasts are right, in fact, and we just got 7 tails in a row. That is a speculative idea, and needs quantification.
Whatever the story, here is the fact: futures prices are not good forecasts (true-measure conditional means) of where interest rates are going. That fact is true not just of Fed Funds futures, but interest rates in general.
Update:
Torsten sends along an updated chart, going further back in history.
Tuesday, May 21, 2019
Clemens on minimum wage
Jeff Clemens offers a "roadmap for navigating recent research" on minimum wages in a nice CATO policy analysis. A review and a doubt.
He discusses the recent claims that minimum wages don't hurt low-skilled people. This is an impressive and readable account of a vast literature. It's not as easy as it seems to evaluate cause and effect in economics. Evidence from small increases in the minimum wage over short time intervals in some locations in good economic times may not tell you the effects of large increases over longer time intervals in all locations in bad economic times.
The "new conventional wisdom," of small effects, Jeff reports, ignores a lot of the more recent work, and especially work that uses "data from individual-level administrative records" rather than "aggregate data and survey data," work that runs "experiments whenever possible," and work that "transparently analyzes compact historical episodes in the U.S. experience" (P. 8)
He discusses the recent claims that minimum wages don't hurt low-skilled people. This is an impressive and readable account of a vast literature. It's not as easy as it seems to evaluate cause and effect in economics. Evidence from small increases in the minimum wage over short time intervals in some locations in good economic times may not tell you the effects of large increases over longer time intervals in all locations in bad economic times.
The "new conventional wisdom," of small effects, Jeff reports, ignores a lot of the more recent work, and especially work that uses "data from individual-level administrative records" rather than "aggregate data and survey data," work that runs "experiments whenever possible," and work that "transparently analyzes compact historical episodes in the U.S. experience" (P. 8)
Thursday, May 9, 2019
Rent Control Poem
"kevinsch" posts an remarkable essay on rent control on a Seattle city council blog (HT Marginal Revolution).
More deeply, he missed the underlying cause of the problem -- building, zoning, and land-use restrictions. Supply meets demand. If builders were allowed to build cheap apartments for modest renters, they would do so. If builders were allowed to build expensive apartments for high-income renters, who then would move out of buildings suitable for low rent apartments, they would do so.
There is a lesson here. Why do our governments, and especially local governments, so often wander into terrible economic policies? The "education" theory says they just don't know basic economics, and don't have any competent policy advice. If they and their staff could just be "educated" things would get better. (And if we could break through all the competing parties who also want to "educate" politicians.) The "interest" theory, more typical among public choice economists, views political outcomes as the result of power, not ideas. Rent control wins when incumbent renters who want to stay put win the political battle over landlords, mobile renters, and potential newcomers, and invoke whatever ideas butter the toast of their cause.
That the city council of Seattle has available such amazingly good policy advice speaks to the latter, sad to say for those of us in the "education" business.
The third view is that ideas still matter at the larger level. A bad idea like rent control requires the asset of the general voter. Yes, incumbent renters who know how to work the system may win the political battle over landlords, property owners, people who want to move to the city and rent, and mobile renters or those not good at working the system, who will lose. But the larger mass of homeowners, condo owners, and non-controlled renters must go along. They don't have a personal interest, other than a general desire to feel good by helping those who face higher rents, so they don't have much reason to study the issue. If the general electorate understood how bad rent control is for their city, and most of the people they want to help, perhaps economic policy would be better. There is hope for ideas.
I’m not an economist, not a landlord, nor a renter. But since we’re having this debate, I went to the UW Library and pulled the literature on rent control so I could understand the issues, the studies, and what the experts conclude. Here’s what I found.
1. Within the community of economists there is broad consensus that rent control is a bad idea. The consensus is on par with the scientific community on climate change, and the medical community on the safety of vaccinations.Given the widespread move to introduce rent controls on the left coast, savor that.
2. There are two documented benefits of rent control: it decreases economic displacement for people living in rent-controlled housing, and it can reduce the volatility of rental pricing in cities where there is sufficient stock of rental housing.
3. There is a very long list of documented harms that rent control causes. It provides a strong disincentive to build more rental housing. It drives landlords to reduce spending on maintaining their units until the quality of the housing has drawn down to the point where it matches the allowed rent. And thus by reducing property values, it reduces property tax revenues. It reduces mobility for renters, causing them to stay in their rent-controlled housing rather than move when a better job or the needs of their family require it. It misallocates the total housing stock by encouraging people to stay in housing that doesn’t match their needs. It encourages rental property owners to convert apartments to condominiums, thereby reducing the rental housing stock. It inevitably leads to a “cluster” of regulations piled on top to try to legislate away all of rent control’s problems. And it doesn’t help the people with the greatest need, but rather the people most capable of gaming the system.It's remarkable that someone who is not an economist could so quickly find all these subtle effects. Yes, most people quickly get that landlords will not keep up apartments, and builders won't build them. But most people don't quickly get the disincentives for renters not to "move when a better job or the needs of their housing require it." Or that it leads not to nirvana for the low income renter, but helps "the people most capable of gaming the system." I would only add that it really hurts the young ambitious person of limited means who wants to move to town to get that upward-mobility job.
4. In many cities with rent control, tenants see annual rent increases at the maximum amount allowed, because landlords understand that if they skip a year they will never catch up.
5. Rent control’s harms can be mitigated in part through an aggressive public/social housing program that creates a large quantity of units using public funds. However, in those places it’s unclear that rent control itself is adding much value beyond the significant value that the public housing program alone delivers.OK, Kevinsch is not an economist so I'll let this pass. The history of aggressive public/social housing programs in US cities are an absolute disaster.
More deeply, he missed the underlying cause of the problem -- building, zoning, and land-use restrictions. Supply meets demand. If builders were allowed to build cheap apartments for modest renters, they would do so. If builders were allowed to build expensive apartments for high-income renters, who then would move out of buildings suitable for low rent apartments, they would do so.
6. As this paper says, rent control “confers its benefits early, and extracts its costs late.” That’s one of the reasons it’s such an attractive policy idea.Well, it confers benefits to renters early. The loss of property value to landlords is instant, but apparently nobody cares about them. The "one time" capital tax is always tempting.
7. As this article puts so well, among rent control advocates there are no rent control failures; there are only bad implementations.Ditto, say, Socialism and Venezuela.
8. And finally, as this research paper suggests, economists have been thorough at convincing themselves that rent control is a bad idea, and inept at convincing anyone else.
This is a gem. And so true. Like, say, tariffs. I wish I knew just how to fix that despite the immense effort and millions of dollars going in to better dissemination of economic ideas.
The essay goes on, and it's worth reading the whole thing.
There is a lesson here. Why do our governments, and especially local governments, so often wander into terrible economic policies? The "education" theory says they just don't know basic economics, and don't have any competent policy advice. If they and their staff could just be "educated" things would get better. (And if we could break through all the competing parties who also want to "educate" politicians.) The "interest" theory, more typical among public choice economists, views political outcomes as the result of power, not ideas. Rent control wins when incumbent renters who want to stay put win the political battle over landlords, mobile renters, and potential newcomers, and invoke whatever ideas butter the toast of their cause.
That the city council of Seattle has available such amazingly good policy advice speaks to the latter, sad to say for those of us in the "education" business.
The third view is that ideas still matter at the larger level. A bad idea like rent control requires the asset of the general voter. Yes, incumbent renters who know how to work the system may win the political battle over landlords, property owners, people who want to move to the city and rent, and mobile renters or those not good at working the system, who will lose. But the larger mass of homeowners, condo owners, and non-controlled renters must go along. They don't have a personal interest, other than a general desire to feel good by helping those who face higher rents, so they don't have much reason to study the issue. If the general electorate understood how bad rent control is for their city, and most of the people they want to help, perhaps economic policy would be better. There is hope for ideas.
Friday, March 22, 2019
Concentration increasing?
Is the US economy getting more concentrated or less? At the aggregate level, more. This is a widely noted fact, leading quickly to calls for more active government moves to break up big companies.
But at the local level, no. Diverging Trends in National and Local Concentration by Esteban Rossi-Hansberg, Pierre-Daniel Sarte, and Nicholas Trachter documents the trend.
They make a concentration measure that is basically the sum of squared market shares, so up means more concentrated and down means less concentrated. This is the average of many different industries and markets.
The average concentration of national markets has gone up. But the concentration of smaller and smaller markets has gone down. More businesses are dividing up county and zip code markets.
Industries differ. This graph does not get a prize for ease of distinguishing the lines, but the two red lines just below zero are manufacturing and wholesale trade, where the industries with really dramatic reductions in local concentration are retail trade, finance insurance and real estate, and services.
What's going on? The natural implication is that the town once had 3 local restaurants, two local banks, and 3 stores. Now it has a McDonalds, a Burger King, a Denny's and an Applebees; a branch of Chase, B of A, and Wells Fargo, and a Walmart, Target, Best Buy, and Costco. National brands replace local stores, increasing the number of local stores.
However, that turns out not to be so obvious.
This graph shows what happens in the diverging industries (those in which national goes up, and local goes down) if you leave out the biggest company. Doing so, lowers the rise of national concentration, because we left out the single most concentrated firm. The lower line however, shows a positive effect. If we leave out the largest national firm, the local markets look more concentrated. If national brands had just replaced local businesses, then when we leave them out, we should see lots of smaller shares. The same thing happens if we leave out the second and third largest.
What's going on? Well, they look at what happens when Wal-Mart comes to town.
The lower line is the effect on concentration in the years before and after the top national firm enters a market. Concentration drops. If, when Wal-Mart came to town, all the exiting firms went under, concentration would rise. The upper line shows you concentration ignoring the largest enterprise. It's unchanged. Either the mom and pop stores do, in fact, stay in business; or new smaller firms enter along with Wal-Mart. The phenomenon is not just the replacement of all smaller businesses by a larger number of national chains.
The paper was presented at the San Francisco Fed "Macroeconomics and Monetary Policy" conference, where I am today. The discussions, by Huiyu Li and François Gourio, were excellent. As with all micro data there is a lot to quibble with. Is a zip code really a market? Much of the data are industry+zip codes with a single firm, both before and (slightly less often) after. Maybe Walmart and other stores drag in customers from other places? And of course, concentration is not the same thing as competition. The SF Fed will, in a week or so, post the conference, papers, and discussions.
But at the local level, no. Diverging Trends in National and Local Concentration by Esteban Rossi-Hansberg, Pierre-Daniel Sarte, and Nicholas Trachter documents the trend.
The average concentration of national markets has gone up. But the concentration of smaller and smaller markets has gone down. More businesses are dividing up county and zip code markets.
What's going on? The natural implication is that the town once had 3 local restaurants, two local banks, and 3 stores. Now it has a McDonalds, a Burger King, a Denny's and an Applebees; a branch of Chase, B of A, and Wells Fargo, and a Walmart, Target, Best Buy, and Costco. National brands replace local stores, increasing the number of local stores.
However, that turns out not to be so obvious.
What's going on? Well, they look at what happens when Wal-Mart comes to town.
The lower line is the effect on concentration in the years before and after the top national firm enters a market. Concentration drops. If, when Wal-Mart came to town, all the exiting firms went under, concentration would rise. The upper line shows you concentration ignoring the largest enterprise. It's unchanged. Either the mom and pop stores do, in fact, stay in business; or new smaller firms enter along with Wal-Mart. The phenomenon is not just the replacement of all smaller businesses by a larger number of national chains.
The paper was presented at the San Francisco Fed "Macroeconomics and Monetary Policy" conference, where I am today. The discussions, by Huiyu Li and François Gourio, were excellent. As with all micro data there is a lot to quibble with. Is a zip code really a market? Much of the data are industry+zip codes with a single firm, both before and (slightly less often) after. Maybe Walmart and other stores drag in customers from other places? And of course, concentration is not the same thing as competition. The SF Fed will, in a week or so, post the conference, papers, and discussions.
Wednesday, March 20, 2019
Less listing
Stocks are fleeing the exchanges in the US. Small and young stocks are disappearing most, with older larger stocks dominating. Less public means more private, not less companies. Companies are more and more financed by private equity, groups of large investors, debt, venture capital and so forth.
This is largely a US phenomenon, which is important for us to figure out what's going on:
What's going on? Doidge, Kahle, Karolyi, and Stulz have some intriguing hypotheses. US business is more and more invested in intellectual capital rather than physical capital -- software, organizational improvements, know-how, not blast furnaces. These, they speculate, are less well financed by issuing shares on the open market, and better by private owners and debt.
Improvements in financial technology such as derivatives allow companies to offload risks without the "agency costs" of equity, and then keep a narrower group of equity investors and more debt financing.
"We argue that the importance of intangible investment has grown but that public markets are not well-suited for young, R&D-intensive companies. Since there is abundant capital available to such firms without going public, they have little incentive to do so until they reach the point in their lifecycle where they focus more on payouts than on raising capital."
I.e. the only reason to go public is for the founders to cash out, and to offer a basically bond-like security for investors. But not to raise capital.
They leave out the obvious question -- to what extent is this driven by regulation? Sarbanes Oxley, SEC, and other regulations and political interference make being a public company in the US a more and more costly, and dangerous, proposition. This helps to answer the question, why in the US.
The move of young, entrepreneurial companies who need financing to grow to private markets, limited to small numbers of qualified investors, has all sorts of downsides. If you worry about inequality, regulations that only rich people may invest in non-traded stocks should look scandalous, however cloaked in consumer protection. But if you can only have 500 investors, they will have to be wealthy. Moving financing from equity to debt and derivatives does not look great from a financial stability point of view.
Our financial system has become remarkably democratized in recent years. Once upon a time only wealthy individuals held stocks, and had access to the superior investment returns they provide. Now index funds, 4501(k) plans are open to everyone, and their pension funds. What will they invest in as listed equity disappears?
A wealth tax, easy to assess on publicly traded stock and much harder to assess on private companies with complex share structures -- especially structures designed to avoid the tax -- will only exacerbate the problem. More moves to regulate the boards and activities of public companies will only exacerbate the problem.
Thursday, March 14, 2019
Competitive deposits?
In its death note to narrow banks (link to Federal Register where you can post comments; previous post), the Fed claimed charmingly that retail deposit rates are fully competitive, so we don't need a narrow bank option to help spread the interest on reserves to deposit rates. In the Fed's view, the fact that banks pay so little compared to reserves just reflects the costs (many of them regulatory!) of servicing retail accounts.
When the Fed Funds rate rises, checking account rates do not. (It's interesting that savings and time deposits do move more quickly, indicating banks face more competition there.) The Fed's story that the spread between checking account rates and federal funds (now IOER) rates reflects costs is very hard to square with this graph -- why should costs and benefits of checking accounts change over time so much, and coincidentally rise exactly one for one with the Federal Funds rate?
Pablo Kurlat, Deposit Spreads and the Welfare Cost of Inflation plots similar data cross sectionally, which lets you estimate the pass through rate better at the expense of the time pattern:
Pablo puts the spread between deposit and federal funds rate on the vertical axis. So, if banks passed through interest rates one for one, the line would be flat. If there were a constant cost, it would be flat but at a higher level. If banks pay the same lousy rate no matter what interest rates are, the curve lies on the 45 degree line. You can see the same general picture.
(Pablo's paper is very nice. He concludes that therefore the "Friedman rule" that interest rates should perpetually be zero, with slight deflation making real rates positive, has yet another thing going for it, that banks are not able to use their market power against us so much.)
Pablo also plots data from different countries:
It's interesting that Sweden and Italy have flatter (more competitive lines). It's really interesting that Argentina lies on the 45 degree line, with no pass through, despite huge inflation-induced interest rates. I would guess that Argentina has a law against paying interest rate on deposits, as the US used to have.
No, it strikes me we have exactly what it seems to be, looking out the window, a heavily regulated not very competitive oligopoly, sort of like airlines 1972.
"Some have argued that the presence of PTIEs could play an important role in raising deposit rates offered by banks to their retail depositors. The potential for rates offered by PTIEs to have a meaningful impact on retail deposit rates, however, seems very low...retail deposit accounts have long paid rates of interest far below those offered on money market investments, reflecting factors such as bank costs in managing such retail accounts and the willingness of retail customers to forgo some interest on deposits for the perceived convenience or safety of maintaining balances at a bank rather than in a money market investment.Here is some data. From "The Deposits Channel of Monetary Policy" by Itamar Drechsler Alexi Savov and Philipp Schnabl, The Quarterly Journal of Economics, 132 (2017)1819–1876:
When the Fed Funds rate rises, checking account rates do not. (It's interesting that savings and time deposits do move more quickly, indicating banks face more competition there.) The Fed's story that the spread between checking account rates and federal funds (now IOER) rates reflects costs is very hard to square with this graph -- why should costs and benefits of checking accounts change over time so much, and coincidentally rise exactly one for one with the Federal Funds rate?
Pablo Kurlat, Deposit Spreads and the Welfare Cost of Inflation plots similar data cross sectionally, which lets you estimate the pass through rate better at the expense of the time pattern:
Pablo puts the spread between deposit and federal funds rate on the vertical axis. So, if banks passed through interest rates one for one, the line would be flat. If there were a constant cost, it would be flat but at a higher level. If banks pay the same lousy rate no matter what interest rates are, the curve lies on the 45 degree line. You can see the same general picture.
(Pablo's paper is very nice. He concludes that therefore the "Friedman rule" that interest rates should perpetually be zero, with slight deflation making real rates positive, has yet another thing going for it, that banks are not able to use their market power against us so much.)
Pablo also plots data from different countries:
It's interesting that Sweden and Italy have flatter (more competitive lines). It's really interesting that Argentina lies on the 45 degree line, with no pass through, despite huge inflation-induced interest rates. I would guess that Argentina has a law against paying interest rate on deposits, as the US used to have.
No, it strikes me we have exactly what it seems to be, looking out the window, a heavily regulated not very competitive oligopoly, sort of like airlines 1972.
Monday, January 21, 2019
Carbon tax update
An interesting question emerged from some discussion surrounding my last carbon tax post. How big will the tax be? The letter says $40 a ton, but then rising. But how far? And in response to what question?
It occurs to me that the two obvious targets lead to radically different answers.
1) The social cost of carbon. This is what economists usually think of as the appropriate Pigouvian tax. In order to pollute, you pay the cost you impose on others by your pollution.
Even the worst-case scenarios now put the cost of carbon emissions at 10% of GDP in the year 2100. Discount that back, divide by all the carbon emitted between now and then, and, you're going to get a pretty small tax.
2) Temperature or quantitative guidelines. Or, "whatever it takes to stop the global temperature from rising more than 1.5 degrees C." Such a tax has to be high enough to basically stop us from using fossil fuels. It would be radically higher, and impose economic costs far higher than 10% of GDP.
When you set a goal of a quantity with no attached price, the price can get pretty high.
I see now some of the back and forth chatter. Anti-carbon types warn that any tax "won't be enough." Now I know what they mean.
So who sets the tax, and on what basis, are important issues we're all fudging over.
Of course, a cynic would take the view that the tax will be set to
3) Maximize government revenue.
Given the behavioral elasticities, that is likely to be a good deal less than #2, as to high a tax will quickly erode the tax base.
PS: to my may CO2-is-not-a-problem commenters. If (or perhaps when) it's all proved to be a hoax, a carbon tax is a lot easier to undo than the alternative regulatory approach!
It occurs to me that the two obvious targets lead to radically different answers.
1) The social cost of carbon. This is what economists usually think of as the appropriate Pigouvian tax. In order to pollute, you pay the cost you impose on others by your pollution.
Even the worst-case scenarios now put the cost of carbon emissions at 10% of GDP in the year 2100. Discount that back, divide by all the carbon emitted between now and then, and, you're going to get a pretty small tax.
2) Temperature or quantitative guidelines. Or, "whatever it takes to stop the global temperature from rising more than 1.5 degrees C." Such a tax has to be high enough to basically stop us from using fossil fuels. It would be radically higher, and impose economic costs far higher than 10% of GDP.
When you set a goal of a quantity with no attached price, the price can get pretty high.
I see now some of the back and forth chatter. Anti-carbon types warn that any tax "won't be enough." Now I know what they mean.
So who sets the tax, and on what basis, are important issues we're all fudging over.
Of course, a cynic would take the view that the tax will be set to
3) Maximize government revenue.
Given the behavioral elasticities, that is likely to be a good deal less than #2, as to high a tax will quickly erode the tax base.
PS: to my may CO2-is-not-a-problem commenters. If (or perhaps when) it's all proved to be a hoax, a carbon tax is a lot easier to undo than the alternative regulatory approach!
Monday, December 3, 2018
Financing innovation
I went to the Financing of Innovation summit at Stanford GSB last Thursday. (Sorry, I can't seem to find a full program online.) Here is a sample of two interesting papers, presented by Amit Seru and Steve Kaplan:
Tuesday, July 24, 2018
Shorter Papers
Ben Leubsdorf at the WSJ does a great job of covering the discussion within economics over too-long papers, picky editing and refereeing, and other issues.
Defensive writing is certainly part of the issue
This isn't necessarily a bad equilibrium.
Defensive writing is certainly part of the issue
“If you want to publish a paper in a top journal, even if you think you have one key insight that can be conveyed succinctly, the referees are not going to take it,” Ms. Finkelstein said.I think Amy would want to clarify this means referees at other journals. Editors are also to blame. We must remember, referees do not take or reject papers, referees advise editors, and it is always the editor's job to make publication decisions.
From an early stage of an academic career, “it becomes pretty clear that you need to check off a pretty long list of items to really convince people that the way you’re interpreting your results is indeed the right way to do it,” Mr. Bazzi said.
..... When you’re trying to anticipate possible criticisms on a controversial topic like the minimum wage, and situate your research in the deep existing literature on the subject, it “quickly adds up to a long paper,” said University of Massachusetts-Amherst economist Arindrajit Dube,....
Mr. Dube said that paper is now in the process of being revised ahead of publication—including acting on a request to make it shorter.However, journals don't encourage length, and there is some sense to the current equilibrium. You write a paper with lots of defensive "what if this what if that." You send it to journals. My typical paper is rejected at 2-3 journals, so by the time it's published I have 6 to 12 reports. My referees are typically thoughtful and diligent, and the paper grows in addressing all of their what-abouts too. Since I haven't been doing detailed empirical work lately, the requests are not nearly as extensive as those authors receive. Then we finally arrive at publication, and the editor says "now cut it down to 40 pages. You can stuff all that into an internet appendix if you like." Which nobody reads.
This isn't necessarily a bad equilibrium.
Friday, July 20, 2018
Nobel Symposium on Money and Banking Day 2
Bernanke
Sadly Ben Bernanke's video and slides are not up on the website. Ben showed some very interesting evidence that the crisis was an unpredictable run, rather than the usual story about predictable defaults resulting from too much credit. Things really did get suddenly a lot worse in September and October 2008. Yes, it's easy to say this is defense against the charge that he should have done more ahead of time. But evidence is evidence, and I find it quite plausible that the relatively small losses in subprime need not have caused such a massive crisis and recession absent a run. Ben says the material is part of a paper he will release soon, so look for it. One can understand that Bernanke is careful about releasing less than perfect drafts of papers and videos.
History
Barry Eichengreen gave a scholarly account of why history matters, especially the great depression, and we should pay more attention to it. (Paper, video.) He aimed squarely at typical economists whose knowledge stopped at Friedman and Schwartz, or perhaps Ben Bernanke's famous non-monetary channels paper, in which bank failures propagated the depression. He emphasized the role of the gold standard and international cooperation or non-cooperation, and warned against facile comparisons of the gold standard experience to today's events and the euro in particular.
Randy Kroszner has a great set of slides and an engaging presentation. He also started on parallels with the great depression, and told well the story of the US default on gold clauses. He closed with a warning about fighting the last war -- particularly apt given the exclusive focus of most of this conference on the events of 2008 -- and on how to start a crisis. In his view when Bank of England Gov Mervyn King said: “We will support Northern Rock." People hear "Northern Rock's in trouble? Run!" Likewise, in my view, speeches by President Bush and Treasury Secretary Paulson did a lot to spark the run in the US.
DSGE
A highlight for me, was the session on DSGE models.
Marty Eichenbaum (video, slides, subsequent paper) gave a nice review of the current status of new Keynesian DSGE models, and how they are developing in reaction to the financial crisis and recession, and the zero bound episode.
Harald Uhlig
Critiques, or more precisely lists of outstanding puzzles and challenges, are often more memorable and novel than positive summaries, and Harald Uhlig delivered a clear and memorable one. (Video, Slides)
Asset prices are a longstanding problem in DSGE models. In typical linearized form, the quantity dynamics are governed by intertemporal substitution, and the asset prices by risk aversion, and neither has much influence on the other. (I learned this from Tom Tallarini.) Rather obviously, our recent recession was all about risk aversion -- people stopped consuming and investing, and tried to move from private to government bonds because they were scared to death, not a sudden attack of thriftiness. There is a lot of current work going on to try to repair this deficiency, but it still lives in the land of extensions of the model rather than the mainstream. Harald also points out a frequently ignored implication of Epstein-Zin utility, the utility index reflects all consumption and anything that enters utility
Financial frictions are blossoming in DSGE models, in two forms: First, HANK or "heterogenous agent" models, which add things like borrowing constraints and uninsurable risks so that the distribution of income matters, and in an eternal quest to make the models work more like static ISLM. Second, in response to the financial crisis (see first day!) stylized models of banking and intermediary finance are showing up. I'm still a little puzzled that the more standard time-varying risk aversion part of macro-finance got ignored, (a plea here) but that is indeed what's going on.
The conundrum, here as elsewhere in DSGE, is that the more people play with the models, the further they get from their founding philosophy: macro models that do talk about monetary policy, (now) financial crises, but that obey the Lucas rules: Optimization, budget constraints, markets, or, more deeply, structures that have some hope of being policy invariant and therefore predictions that will survive the Lucas critique. Already, many ingredients such as Calvo pricing are convenient parables, but questionably realistic as policy-invariant.
Harald points out that since most of the frictions are imposed in a rather ad-hoc manner, neither will they be policy-invariant. This is a deeper and more realistic point than commonly realized. Every time market participants hit a "friction," they tend to innovate a way around that friction so it doesn't hurt them next time. Regulation Q on interest rates was once a "friction," and then the money market fund was invented. The result is too often "chicken papers:"
The understandable trouble is, if you try to microfound every single friction from Deep Theory -- just why it is that credit card companies put a limit on how much you can borrow, in terms of asymmetric information, moral hazard, and so forth -- the audience will be asleep long before you get to the data. Also, as we saw in day 1, there is (to put it charitably) a lot of uncertainty in just how contract or banking theory maps to actual frictions. I think we're stuck with ad-hoc frictions, if you want to go that route.
Harald's next point is, I think, his most devastating, as it describes a huge hole in current models that is not (unlike the last two) a point of immense current research effort. The Phillips curve and inflation are the central point of the New Keynesian DSGE model -- and a disaster.
The Phillips curve is central. The point of the model is for monetary policy to have output effects. Money itself has (rightly) disappeared in the model, so the only channel for monetary policy to work is via the Phillips curve. Interest rates change inflation, and inflation causes output changes. No surprise, it is very hard for that model to produce anything like the last recession out of small changes in inflation. (I have to agree here with the premise of the financial frictions view -- if you want your model to produce the last recession, other than by one huge shock, the model needs something like a financial crisis.)
The Phillips curve in the data is well known
The point of the slide, in simpler form: The standard Phillips curve is
inflation today = beta x expected inflation next year + kappa x output gap + shock
Essentially all inflation is accounted for by the shock. The model is basically silent about the source of inflation. Looking at the model as a whole, not just one equation, Neither monetary policy shocks nor changes in rules accounts for any significant amount of inflation.
I made a similar graph recently. Use the standard three equation model
Now, use actual data on output y, inflation pi, and interest rate i, to back out the shocks v. Turn off the monetary policy shock vi = 0. Solve the model and plot the data -- what would have happened if the Fed had exactly followed the Taylor rule?
- Data: no Phillips-Curve tradeoff.
- QDSGE: don’t account for inflation with monetary policy shocks.
- The NK / Phillips-Curve-based NK QDSGE models may thus provide a poor guide for monetary policy.
Wait, you ask, what about Marty Eichenbaum's pretty graphs, such as this one, showing the effects of a monetary policy shock?
Last, standard new Keyensian DSGE models have strong "Fisherian" properties. In response to long lasting or expected interest rate rises, inflation goes up. More on this later.
Ellen McGrattan
Ellen stole the show. (Slides.) Take a break, and watch the video. She manages to be hilarious and incisive. And unlike the rest of us, she didn't try to sheohorn a two hour lecture into her 15 minutes.
Her central points. First, like Harald, she points out that the models are driven by large shocks with less and less plausible structural interpretation, and thus further from the Lucas critique solution than once appeared to be the case. The shocks are really "wedges," deviations from equilibrium conditions of the model with unknown sources
What to do? Focus on rules and institutions. This is a deep point. Even DSGE modelers, in the desire to speak to policy makers, often adopt the static ISLM presumption that policy is about actions, about decisions, whether to raise or lower the funds rate. The other big Lucas point is that we should think about policy in terms of rules and institutions, not just actions.
Monetary policy and ELB
Stephanie Schmitt-Grohé (slides, video) talked about the Fisherian possibility -- that raising interest rates raises inflation. New-Keynesian DSGE models, with rational expectations, have this property, especially for permanent or preannounced interest rate increases, and when at zero interest rates or otherwise in a passive regime where interest rates do not react more than one for one with inflation. She and Martin Uribe have been advocating this possibility as a serious proposal for Europe and Japan that want to raise inflation.
She presented some nice evidence that permanent increases in interest rates do increase inflation -- and right away, not just in the long run.
Mike Woodford. (slides, video) gave a dense talk (37 slides, 20 minutes) on policy at the lower bound. During the ELB, central banks moved from interest rates to asset purchases and forward guidance. Mike asks,
To what extent does this mean that the entire conceptual framework of monetary stabilization policy needs to be reconsidered, for a world in which ELB might well continue periodically to bind?In classic form, Mike sets the question up as a Ramsey problem. Given a DSGE model, what is the optimal policy, given that interest rates are occasionally constrained? He derives from that problem a price level target. The price level target works, intuitively, by committing the central bank to a period of extra inflation after the zero bound ends. It is a popular form of forward guidance. The innovation here is to derive that formally as an optimal policy problem.
Mike's price level target is stochastic, changing optimally over time to respond to shocks. I'm a little skeptical that the central bank can observe and understand such shocks, especially given the above Uhlig-McGrattan discussion about the nature of shocks. Also, as I emphasize in comments, I'm dubious about the great power of promises of what the central bank will do in the far future to stimulate output today. I'm a fan of price level targets, but on both sides, not just as stimulus, but for utterly different reasons.
Mike takes on rather skeptically the common alternative -- quantitative easing, asset purchases during the time of the bound. He points out that to work, people have to believe that the increase in money is permanent, and won't be quickly withdrawn when the zero bound is over. As evidence, he points to Japan:
Similarly, he likes the price level target over forward guidance -- speeches in place of action -- as it is a more credible commitment to do things ex-post that the bank may not wish to do ex-post.
Finally, he addresses the puzzles of new Keynesian models at the zero bound -- forward guidance has stronger effects the further in the future is the promise; effects get larger as prices get less sticky, and so on. He argues that models should replace rational expectations with a complex k-step iterated expectations rule.
Me.
Video, slides from Sweden, slides from my webpage, written version. I covered this in a previous blog post, so won't repeat it all. I put a lot of effort in to it, and it summarizes a lot of what I've been doing in 15 minutes flat, so I recommend it (of course). It also offers more perspective on above points by Mike and Stephanie. My favorite line, referring to Mike's push for irrational expectations is something close to
"I never thought we would come to Sweden, that I would be defending the basic new-Keynesian program, and that Mike Woodford would be trying to tear it down. Yet here we are. Promote the fiscal equation from the footnotes and you can save the rest."Emi Nakamura
Poor Emi had to go last in an exhausting conference of jet-lagged participants. She did a great job (video, slides) covering a century of monetary history and monetary ideas clearly and transparently. These are great slides to use for an undergraduate or MBA class on monetary policy, as well. An abbreviated list:
- Gold standard
- Seasonal variation in interest rates under the gold standard; money demand shocks
- Money demand shocks in the 1980s -- how the supposedly "stable" V in MV=PY fell apart when the Fed pushed on M.
- Theoretical instability / indeterminacy of interest rate targets
- The switch to interest rate targets and corridors in operating procedures
- The (near-miraculous) success of inflation targets
- Taylor rules and other theory of determinate inflation under interest rate targets
- How is it "monetary economics" without money?
- Why did immense QE not cause inflation?
Postlude
Monday featured two panels, Macroeconomic research and the financial crisis: A critical assessment, with Annette Vissing-Jørgensen, Luigi Zingales, Nancy Stokey, and Robert Barro ; and Banking and finance research and the financial crisis: A critical assessment with Kristin Forbes, Ricardo Reis, Amir Sufi, and Antoinette Schoar.
Perhaps it's in the nature of panels, but I found these a disappointment, especially compared to the stellar presentations in the main conference. Also I think it would have been better to allow more (any, really) audience questions; the whole conference was a bit disappointing for lack of general discussion, especially with such a stellar group.
In particular, Luigi led by excoriating the profession for not paying attention to housing problems and financial crises. I thought this a bit unfair and simultaneously short-sighted. He singled out monetary economics textbooks, including Mike Woodford's, for omitting financial crises. Well, Mike omitted asteroid impacts too. It isn't a book about financial crises. And, after lamabasting all of us, he said not one word about events since 2009. What are we missing now? I had to stand up and ask that rude question, again suggesting that perhaps we are all not listening to Ken Rogoff this time. Annette went on to ask something like "don't you Chicago people believe in any regulation at all," and the respondents were too polite to say what an unproductive question that is and just move on.
Again, I offer apologies to authors and discussants I didn't get to. The whole thing was memorable, but there is only so much I can blog! Do go to the site and look at the other sessions, according to your interests.
Subscribe to:
Posts (Atom)