This post by Nick Rowe:
Returning from a long weekend's canoeing, I remembered one of the main reasons I am a sticky-price macroeconomist. I checked to see if the economy had imploded while I was not worrying about it. April CPI up 0.1% from March; CAD/USD down about 2%; CAD/EUR up about 2%; TSX down around 3%; S&P500 down around 6%; oil down around 3%, etc.
You get the picture. The exact numbers don't matter. The CPI moved less in one whole month than all the other numbers moved in my 4-day weekend. A whole order of magnitude less. The CPI is a boring number. Anybody can predict next month's CPI within a 1% range; nobody can predict next month's exchange rates, oil prices, or stock prices with that degree of accuracy.
But what does this tell us?
My long weekend was a time of big changes in demands and supplies of financial assets, and the prices of all those financial assets relative to each other changed a lot in 4 days. But nothing seemed to have changed when I took my Loonies to the local supermarket. And I bet nothing much changed when Europeans and Americans took their Euros and US dollars to their local supermarkets, buying bread and milk like I did, even though the Loonie, Euro, and US dollar had all changed relative to each other. (OK, gas prices were about 2% lower when I returned, but that might be because they always seem to go up at the beginning of a long weekend.)
We don't usually think of the CPI as an asset price, but it is. Or rather, its inverse is. The CPI is the price of a basket of goods in terms of money, so 1/CPI is the price of money in terms of goods. Money is a financial asset, so 1/CPI is the price of a financial asset. TSX/CPI is the price of another financial asset, the TSX basket of shares, in terms of goods. CAD/USD is the price of one money in terms of another money. 1/CPI is much more predictable, at a one month horizon, than TSX/CPI or CAD/USD. About one order of magnitude more predictable.
This fact tells me that most of the prices that make up the CPI are sticky. I can't make sense of it any other way. I don't like the assumption of sticky prices; I hate that assumption, because it is assuming something we ought to be explaining. But until I hear a satisfactory explanation for why the monthly CPI is so predictable, I don't have a better alternative.
Why are prices sticky? Is it because consumers expect predictability?
1. Twenty years ago I paid $30 for a pair of sneakers. Truthfully, it would be hard to persuade me to pay any more today. Fortunately, for technological progress and offshoring, sneaker prices have been able to stay around the $30 range. Unfortunately, they're not quite the same quality sneakers. Same with shirts - I'd be more than a little upset to pay more than $20 for a shirt.
2. 5 years ago we spent the weekend at the Hyatt Chesapeake Resort. We probably paid about $250 per night for the room for a stay in October. These days we go during the off season where rooms can be had for $99 per night. Even today I would be aghast at paying more than $250 per night. Prices in fact have gone up to $350 per night during the peak season (late spring - summer) but at in late fall can still be had for $250.
Or perhaps it's not expectations that are sticky and maybe it's just that I'm cheap.
Thursday, May 27, 2010
Wednesday, May 26, 2010
Models and reality
The dustup between Mark Thoma and David Andolfatto summarized by Rajiv Sethi is perhaps more symptomatic of the divide between - at extreme risk of too much simplification - "new" macroeconomists and "old" macroeconomists. The macroeconomists of my generation were taught DSGE models. Facts were "stylized" facts, i.e. first and second moments of "key" economic variables such as GNP, investment and consumption. During my entire 6 years at graduate school things like institutional details and historical events that may have affected the economy were laid aside or treated as not being "relevant" to the model. Economies were frictionless and markets always cleared. Sure, some frictions were eventually introduced but perhaps the biggest elephant in the room was that the curriculum cultivated us with a certain attitude that:
1) There are those who can build DSGE models and there are those who can't.
2) All partial equilibrium models can be dismissed off hand.
3) All structural equation models are completely irrelevant especially those not based on DSGE models. (IS-LM or Keynesian "cross" models are definitely in this category.)
4) Any paper that does not present a model can be dismissed - this included narratives as well as historical papers.
Perhaps the economists who fail to understand history will be doomed to repeat them?
1) There are those who can build DSGE models and there are those who can't.
2) All partial equilibrium models can be dismissed off hand.
3) All structural equation models are completely irrelevant especially those not based on DSGE models. (IS-LM or Keynesian "cross" models are definitely in this category.)
4) Any paper that does not present a model can be dismissed - this included narratives as well as historical papers.
Perhaps the economists who fail to understand history will be doomed to repeat them?
Tuesday, May 25, 2010
What I've been reading
Some non-fiction stuff:
1. John Boslough, Stephen Hawking's Universe: short and easy read though somewhat dated. Having read several cosmology type books I was able to follow the arguments and I still think that it's a good introduction.
2. George Smoot and Keay Davidson, Wrinkles in Time. Entertaining though some of the technical details eluded me especially the monopole and dipole discussion. I especially liked the figures especially the one labeled Big Bag: The Evolving Universe.
3. F. David Peat, Einstein's Moon: Bell's Theorem and the Curious Quest for Quantum Reality. The experimental details on Bell's Theorem and its extension of the EPR paradox was well presented. Short and clarifying. Enjoyed it.
4. Denise Shekerjian, Uncommon Genius. Disappointing although I'd have to say that this is a tough subject - to find what factors determine creative impulses of MacArthur Fellows. Ultimately, I think I may have been happier if this were short biographies of the fellows she interviewed and to leave the reader to tie the links together. Alternatively, something in the format of Three Scientists and their Gods might have worked better.
5. Roger S. Jones, Physics as Metaphor. Disappointing. I mostly skimmed it. The book seemed rather rambling and its basic thrust was in the school of "reality is what we make of it". I'm sympathetic to this view though I think the presentation of the ideas weren't as entertaining as I would have liked them to be.
6. Fritjof Capra, The Turning Point. This was a book that 20 years ago I would have whole heartedly embraced. While jaded and indoctrinated by free markets, deep down I still cling to the belief that there are many things that cannot be looked at piecemeal (Capra calls it the Cartesian/Newtonian framework) and that the systems view should be adopted. I think there has been some shift to this (though I suspect not as much as he would like) especially with the holistic approaches to healing and health. But (gasp!) not economices. Yeah, one day I would like to see a book called Holistic Economics or the Zen of Economic Systems.
1. John Boslough, Stephen Hawking's Universe: short and easy read though somewhat dated. Having read several cosmology type books I was able to follow the arguments and I still think that it's a good introduction.
2. George Smoot and Keay Davidson, Wrinkles in Time. Entertaining though some of the technical details eluded me especially the monopole and dipole discussion. I especially liked the figures especially the one labeled Big Bag: The Evolving Universe.
3. F. David Peat, Einstein's Moon: Bell's Theorem and the Curious Quest for Quantum Reality. The experimental details on Bell's Theorem and its extension of the EPR paradox was well presented. Short and clarifying. Enjoyed it.
4. Denise Shekerjian, Uncommon Genius. Disappointing although I'd have to say that this is a tough subject - to find what factors determine creative impulses of MacArthur Fellows. Ultimately, I think I may have been happier if this were short biographies of the fellows she interviewed and to leave the reader to tie the links together. Alternatively, something in the format of Three Scientists and their Gods might have worked better.
5. Roger S. Jones, Physics as Metaphor. Disappointing. I mostly skimmed it. The book seemed rather rambling and its basic thrust was in the school of "reality is what we make of it". I'm sympathetic to this view though I think the presentation of the ideas weren't as entertaining as I would have liked them to be.
6. Fritjof Capra, The Turning Point. This was a book that 20 years ago I would have whole heartedly embraced. While jaded and indoctrinated by free markets, deep down I still cling to the belief that there are many things that cannot be looked at piecemeal (Capra calls it the Cartesian/Newtonian framework) and that the systems view should be adopted. I think there has been some shift to this (though I suspect not as much as he would like) especially with the holistic approaches to healing and health. But (gasp!) not economices. Yeah, one day I would like to see a book called Holistic Economics or the Zen of Economic Systems.
Fiction reading
Catching up on some regular fiction:
1. Paul West, The Women of Whitechapel and Jack the Ripper: I had to read it just to see what his theory was. The prose didn't really hold me.
2. Jodi Picoult, Plain Truth. I'd have to say that this was a pretty good book and this was my first Picoult which makes me want to try another.
3. Michael Cunningham, At Home at the End of the World. Wonderful writing but the lack of differences among the voices sometimes confused me. I guess I could psychoanalyze this and say that it is a multiple selves version of one character but I enjoyed the book.
1. Paul West, The Women of Whitechapel and Jack the Ripper: I had to read it just to see what his theory was. The prose didn't really hold me.
2. Jodi Picoult, Plain Truth. I'd have to say that this was a pretty good book and this was my first Picoult which makes me want to try another.
3. Michael Cunningham, At Home at the End of the World. Wonderful writing but the lack of differences among the voices sometimes confused me. I guess I could psychoanalyze this and say that it is a multiple selves version of one character but I enjoyed the book.
Monday, May 24, 2010
Robert Kaplan on Afghanistan
Robert Kaplan writes:
“Afghanistan was a cakewalk in 2001 and 2002,” says Sarah Chayes, former special adviser to McChrystal’s headquarters. “We started out with a country that hated the Taliban and by 2009 were driving people back into the arms of the Taliban. That’s not fate. That’s poor policy.” We enabled an administration, led by Hamid Karzai, that is less a government than a protection racket, in which bribery is the basis of a whole chain of transactions, from small sums paid to criminals at roadblocks in the south of the country to tens of millions of dollars smuggled out of the Kabul airport by government ministers. The myth is that the absence of governance in Afghanistan creates a vacuum in which the Taliban thrive. But the truth, as Chayes explains, is the opposite. Karzai governs everywhere in the revenue belt, synonymous with Pashtunistan, in the south and east of the country: the Taliban succeed in these very places, not because of no governance but because of corrupt and abusive governance.
Referring to the evolution of the former mujahideen commanders into gangster-oligarchs under Karzai, an Afghan analyst, Walid Tamim, told me: “Warlords like Rabbani, Fahim, Sayyaf, and Dostum have all been empowered by Karzai and the U.S. government. Why is [Taliban leader] Mullah Omar any worse than these guys?” Ashraf Ghani, the country’s finance minister from 2002 to 2004, explained: “The core threat we all face is the Afghan government itself. About two-thirds of revenue is lost to abuse. This isn’t like corruption in Indonesia, where money is stolen but things still get built; here it is all looted, because the warlords are insecure about what may come next in Afghan politics.” Even as American officers talk publicly in bland clichés about partnering with and improving the performance of the Karzai government, the grim reality of Afghan public life is distinguished by corruption, criminality, and poverty.
I knew and wrote about Karzai in the 1980s, when he was a representative in Peshawar of the pro-Western mujahideen faction of Sibghatullah Mojaddedi. Mojaddedi had very little military presence inside Afghanistan; he and Karzai were no threat to anybody. Karzai had impressed me as personable, enlightened, sensitive, and, now that I think about it over the distance of time, weak. I genuinely liked him. But alas, he is said to be bored by actual governance. As Ghani points out, “He is not an organization man with the requisite management abilities,” and thus he lacks the skill to build a popular power base like the one the late Afghan Communist leader Babrak Karmal was able to build in the late 1970s and early 1980s, or even like the one the Soviet puppet Najibullah built later on. And without a power base of his own, and with the Americans distracted since 2003 by Iraq, Karzai has had few others to rely on but the warlords and his own knee-deep-in-graft family.
And on ethnic and tribal loyalties:
What does it mean to work with the tribes, Churchill-style; what does it take to overcome the geographical and human terrain here? The story of Colonel Chris Kolenda, of Omaha, Nebraska, is instructive. Kolenda, a West Point graduate with the sharp-eyed, comforting manner of a family physician, commanded the 1st Squadron of the 91st Cavalry from May 2007 to July 2008 in northeastern Afghanistan, on the border with Pakistan. When Kolenda’s 800-soldier battalion arrived, armed violence was endemic. Coalition headquarters in Kabul blamed a Pakistan-based insurgency. “The conventional wisdom was wrong,” Kolenda told me. “Almost all of the insurgents were locals who fought for a whole variety of reasons: they were disgusted with ISAF, as well as the government in Kabul; their fathers had fought the Soviets and now the sons were fighting the new foreigners.”
Then there was the “psychodrama of interethnic and clan frictions,” abetted by the fractured mountainous landscape. The area was populated by Nuristanis, Kohistanis, and Pashtuns, all of whom harbored disdain for the Gujars, migrant farm workers from over the border, who, in their eyes, were “not real Afghans.” (So much for the argument that there is no Afghan national identity.) The Nuristanis, in turn, were divided into the Kata, Kom, Kushtowz, and Wai clans. The Kom were split into hostile and well-armed groups whose current divisions stemmed from the war against the Soviets in the 1980s, when some of the Kom backed the radical forces of Gulbuddin Hekmatyar, known as the HIG, or Hezb-i-Islami-Gulbuddin, and other Kom sub-clans were loyal to the moderate National Islamic Front of Afghanistan. The Kata, meanwhile, were generally loyal to the Lashkar-e-Taiba (“Army of the Righteous”), which carried out major attacks against India from bases in Pakistan. The Pashtuns themselves were divided in some cases, on account of blood feuds, into five elements.
Kolenda apologized to me for “getting down in the weeds,” but explained that until he’d learned who was who, and who was fighting whom, his battalion couldn’t make progress and escape the cycle of ferocious firefights that had characterized the first three months of its deployment. “People were often giving us tips about bad guys who weren’t really bad guys, but simply people from another faction with whom the tipster had a score to settle.”
Interesting and thoughtful article throughout.
“Afghanistan was a cakewalk in 2001 and 2002,” says Sarah Chayes, former special adviser to McChrystal’s headquarters. “We started out with a country that hated the Taliban and by 2009 were driving people back into the arms of the Taliban. That’s not fate. That’s poor policy.” We enabled an administration, led by Hamid Karzai, that is less a government than a protection racket, in which bribery is the basis of a whole chain of transactions, from small sums paid to criminals at roadblocks in the south of the country to tens of millions of dollars smuggled out of the Kabul airport by government ministers. The myth is that the absence of governance in Afghanistan creates a vacuum in which the Taliban thrive. But the truth, as Chayes explains, is the opposite. Karzai governs everywhere in the revenue belt, synonymous with Pashtunistan, in the south and east of the country: the Taliban succeed in these very places, not because of no governance but because of corrupt and abusive governance.
Referring to the evolution of the former mujahideen commanders into gangster-oligarchs under Karzai, an Afghan analyst, Walid Tamim, told me: “Warlords like Rabbani, Fahim, Sayyaf, and Dostum have all been empowered by Karzai and the U.S. government. Why is [Taliban leader] Mullah Omar any worse than these guys?” Ashraf Ghani, the country’s finance minister from 2002 to 2004, explained: “The core threat we all face is the Afghan government itself. About two-thirds of revenue is lost to abuse. This isn’t like corruption in Indonesia, where money is stolen but things still get built; here it is all looted, because the warlords are insecure about what may come next in Afghan politics.” Even as American officers talk publicly in bland clichés about partnering with and improving the performance of the Karzai government, the grim reality of Afghan public life is distinguished by corruption, criminality, and poverty.
I knew and wrote about Karzai in the 1980s, when he was a representative in Peshawar of the pro-Western mujahideen faction of Sibghatullah Mojaddedi. Mojaddedi had very little military presence inside Afghanistan; he and Karzai were no threat to anybody. Karzai had impressed me as personable, enlightened, sensitive, and, now that I think about it over the distance of time, weak. I genuinely liked him. But alas, he is said to be bored by actual governance. As Ghani points out, “He is not an organization man with the requisite management abilities,” and thus he lacks the skill to build a popular power base like the one the late Afghan Communist leader Babrak Karmal was able to build in the late 1970s and early 1980s, or even like the one the Soviet puppet Najibullah built later on. And without a power base of his own, and with the Americans distracted since 2003 by Iraq, Karzai has had few others to rely on but the warlords and his own knee-deep-in-graft family.
And on ethnic and tribal loyalties:
What does it mean to work with the tribes, Churchill-style; what does it take to overcome the geographical and human terrain here? The story of Colonel Chris Kolenda, of Omaha, Nebraska, is instructive. Kolenda, a West Point graduate with the sharp-eyed, comforting manner of a family physician, commanded the 1st Squadron of the 91st Cavalry from May 2007 to July 2008 in northeastern Afghanistan, on the border with Pakistan. When Kolenda’s 800-soldier battalion arrived, armed violence was endemic. Coalition headquarters in Kabul blamed a Pakistan-based insurgency. “The conventional wisdom was wrong,” Kolenda told me. “Almost all of the insurgents were locals who fought for a whole variety of reasons: they were disgusted with ISAF, as well as the government in Kabul; their fathers had fought the Soviets and now the sons were fighting the new foreigners.”
Then there was the “psychodrama of interethnic and clan frictions,” abetted by the fractured mountainous landscape. The area was populated by Nuristanis, Kohistanis, and Pashtuns, all of whom harbored disdain for the Gujars, migrant farm workers from over the border, who, in their eyes, were “not real Afghans.” (So much for the argument that there is no Afghan national identity.) The Nuristanis, in turn, were divided into the Kata, Kom, Kushtowz, and Wai clans. The Kom were split into hostile and well-armed groups whose current divisions stemmed from the war against the Soviets in the 1980s, when some of the Kom backed the radical forces of Gulbuddin Hekmatyar, known as the HIG, or Hezb-i-Islami-Gulbuddin, and other Kom sub-clans were loyal to the moderate National Islamic Front of Afghanistan. The Kata, meanwhile, were generally loyal to the Lashkar-e-Taiba (“Army of the Righteous”), which carried out major attacks against India from bases in Pakistan. The Pashtuns themselves were divided in some cases, on account of blood feuds, into five elements.
Kolenda apologized to me for “getting down in the weeds,” but explained that until he’d learned who was who, and who was fighting whom, his battalion couldn’t make progress and escape the cycle of ferocious firefights that had characterized the first three months of its deployment. “People were often giving us tips about bad guys who weren’t really bad guys, but simply people from another faction with whom the tipster had a score to settle.”
Interesting and thoughtful article throughout.
Sunday, May 23, 2010
A financial blog post round up
A round up of some blog posts I had been sitting on and racking my brains on how to intelligently discuss them but unable to come up with anything so I thought I'd just list them here with some short comments:
1. I had previously wondered whether the U.S. would experience a lost decade here and here and Krugman asserts that this is a looming possibility:
Recent data don’t suggest that America is heading for a Greece-style collapse of investor confidence. Instead, they suggest that we may be heading for a Japan-style lost decade, trapped in a prolonged era of high unemployment and slow growth.
Tim Duy:
To summarize, the Fed believes we are facing another threat to demand, either via financial or real trade linkages, at a time when lending activity continues to fall, suggesting that monetary policy is too tight to begin with. But the Fed stance is to believe that monetary policy is on the verge of being too loose, and, if anything, planning needs to be made to tighten policy. At the same time, Fed policymakers also believe fiscal policy needs to turn toward tightening as well. Meanwhile, unemployment hovers just below 10%, nor is it expected to decline rapidly, and inflation continues to trend downward.
All of which together suggests that the Fed's policy stance is seriously out of whack with policymaker's interpretation of actual and potential economic developments. And I have trouble explaining the disconnect.
One of the things we tried to do a few months ago was to rebalance our portfolio based on what we thought the likley outcome in the next 10 years would be. Our thought was that the Fed/Treasury would try to inflate its way out of debt (somewhat) and inflation would rise. Even if it did not, the increase in debt would place upward pressure on interest rates. As such we moved most of our savings out of equities into TIPS. Where does the current commentary leave us?
As Krugman points out, interest rates have actually fallen due to the Greek crisis and the deflation is now a possible scenario (again!). While my longer term outlook for interest rates remains the same, the likelihood of this happening is decreasing - in other words, I have to revise downward my prior on the probability of inflation being likely in the medium term. Unfortunately, with the battering the equity and bond markets are taking there doesn't seem to be any safe sanctuary for our retirement savings.
2. That ratings agencies amplify crises and really have no role in regulation was something I had not thought about before:
During the boom, early rating downgrades would help to dampen euphoric expectations and reduce private short-term capital flows which have repeatedly been seen to fuel credit booms and financial vulnerability in the capital-importing countries. By contrast, if sovereign ratings had no market impact, they would be unable to smooth boom–bust cycles. Worse, if sovereign ratings lag rather than lead financial markets, but have a market impact, improving ratings would reinforce euphoric expectations and stimulate excessive capital inflows during the boom whereas during the bust, downgrading might add to panic among investors, driving money out of the country and sovereign yield spreads up. If guided by outdated crisis models, sovereign ratings would fail to provide early warning signals ahead of a currency crisis, which again might reinforce herd behaviour by investors.
The clincher: Rather than to establish yet another rating agency, the appropriate policy response will be a thorough revision, in fact exclusion, of the role of sovereign ratings in prudential regulation and even in internal industry guidelines.
3. However, the fact that derivatives can be destabilizing is something that I had been aware of and is now part of the popular culture thanks to Michael Lewis in the Big Short. Ideally, the existence of CDS would have allowed the shorts to make markets more efficient by signalling that they thought the underlying mortgage securities were over-priced. The fact that the shorts were able to make billions by buying so many CDSs (in effect taking the opposite side of the securitization) without affecting the price of the CDOs that were being sold to investors seem to me to be a failure of information transmission - the markets aren't transparent enough. If I knew that someone was willing to take a large position on the opposite side of my transaction then this would give me pause.
Perhaps the problem is so much derivatives per se but the lack of understanding in the markets and how they are interlinked as well as largely the fact that these transactions are not transparent. Could the existence of a more liquid and transparent market in CDSs and CDOs/mortgage securities been able to prevent the bubble? Hardly - the existence of put options for S&P and other stocks have not been able to prevent a bubble and it's hard to believe that this might be a self-regulating mechanism.
4. A very articulate response from Kocherlakota on why macroeconomic models failed us. The thrust is that macro models and the act of modeling are very segmented. Labor market frictions, price frictions and financial and asset market frictions are features of some macro models and not all models. In other words, it is hard to build a coherent macro model with all the frictions in at once. A model like this would be more realistic.
However, incorporating the financial sector is extremely hard. Perhaps it is made all the more harder by the fact that economists rarely try to work together to build these models and they rarely share code. Perhaps this is because of the publish or perish mentality - that if I share my code with you, you may be able to beat me to getting published just by tweaking my code a little. I don't know how relevant this is but the fact that different economists are working on different aspects of economic frictions and not coming together to try to build a large scale model says several things:
a) They do not believe that all frictions are created equal - my friction is bigger than your friction
b) Large scale DSGE models are impossible perhaps due to computational constraints and perhaps it is somewhat related the stigma attached to large scale macroeconometric modeling that took place in the 1970s (and its subsequent failure)
c) Economists actually believe that it is better to compete amongst themselves to build better models than to cooperate to build better models.
1. I had previously wondered whether the U.S. would experience a lost decade here and here and Krugman asserts that this is a looming possibility:
Recent data don’t suggest that America is heading for a Greece-style collapse of investor confidence. Instead, they suggest that we may be heading for a Japan-style lost decade, trapped in a prolonged era of high unemployment and slow growth.
Tim Duy:
To summarize, the Fed believes we are facing another threat to demand, either via financial or real trade linkages, at a time when lending activity continues to fall, suggesting that monetary policy is too tight to begin with. But the Fed stance is to believe that monetary policy is on the verge of being too loose, and, if anything, planning needs to be made to tighten policy. At the same time, Fed policymakers also believe fiscal policy needs to turn toward tightening as well. Meanwhile, unemployment hovers just below 10%, nor is it expected to decline rapidly, and inflation continues to trend downward.
All of which together suggests that the Fed's policy stance is seriously out of whack with policymaker's interpretation of actual and potential economic developments. And I have trouble explaining the disconnect.
One of the things we tried to do a few months ago was to rebalance our portfolio based on what we thought the likley outcome in the next 10 years would be. Our thought was that the Fed/Treasury would try to inflate its way out of debt (somewhat) and inflation would rise. Even if it did not, the increase in debt would place upward pressure on interest rates. As such we moved most of our savings out of equities into TIPS. Where does the current commentary leave us?
As Krugman points out, interest rates have actually fallen due to the Greek crisis and the deflation is now a possible scenario (again!). While my longer term outlook for interest rates remains the same, the likelihood of this happening is decreasing - in other words, I have to revise downward my prior on the probability of inflation being likely in the medium term. Unfortunately, with the battering the equity and bond markets are taking there doesn't seem to be any safe sanctuary for our retirement savings.
2. That ratings agencies amplify crises and really have no role in regulation was something I had not thought about before:
During the boom, early rating downgrades would help to dampen euphoric expectations and reduce private short-term capital flows which have repeatedly been seen to fuel credit booms and financial vulnerability in the capital-importing countries. By contrast, if sovereign ratings had no market impact, they would be unable to smooth boom–bust cycles. Worse, if sovereign ratings lag rather than lead financial markets, but have a market impact, improving ratings would reinforce euphoric expectations and stimulate excessive capital inflows during the boom whereas during the bust, downgrading might add to panic among investors, driving money out of the country and sovereign yield spreads up. If guided by outdated crisis models, sovereign ratings would fail to provide early warning signals ahead of a currency crisis, which again might reinforce herd behaviour by investors.
The clincher: Rather than to establish yet another rating agency, the appropriate policy response will be a thorough revision, in fact exclusion, of the role of sovereign ratings in prudential regulation and even in internal industry guidelines.
3. However, the fact that derivatives can be destabilizing is something that I had been aware of and is now part of the popular culture thanks to Michael Lewis in the Big Short. Ideally, the existence of CDS would have allowed the shorts to make markets more efficient by signalling that they thought the underlying mortgage securities were over-priced. The fact that the shorts were able to make billions by buying so many CDSs (in effect taking the opposite side of the securitization) without affecting the price of the CDOs that were being sold to investors seem to me to be a failure of information transmission - the markets aren't transparent enough. If I knew that someone was willing to take a large position on the opposite side of my transaction then this would give me pause.
Perhaps the problem is so much derivatives per se but the lack of understanding in the markets and how they are interlinked as well as largely the fact that these transactions are not transparent. Could the existence of a more liquid and transparent market in CDSs and CDOs/mortgage securities been able to prevent the bubble? Hardly - the existence of put options for S&P and other stocks have not been able to prevent a bubble and it's hard to believe that this might be a self-regulating mechanism.
4. A very articulate response from Kocherlakota on why macroeconomic models failed us. The thrust is that macro models and the act of modeling are very segmented. Labor market frictions, price frictions and financial and asset market frictions are features of some macro models and not all models. In other words, it is hard to build a coherent macro model with all the frictions in at once. A model like this would be more realistic.
However, incorporating the financial sector is extremely hard. Perhaps it is made all the more harder by the fact that economists rarely try to work together to build these models and they rarely share code. Perhaps this is because of the publish or perish mentality - that if I share my code with you, you may be able to beat me to getting published just by tweaking my code a little. I don't know how relevant this is but the fact that different economists are working on different aspects of economic frictions and not coming together to try to build a large scale model says several things:
a) They do not believe that all frictions are created equal - my friction is bigger than your friction
b) Large scale DSGE models are impossible perhaps due to computational constraints and perhaps it is somewhat related the stigma attached to large scale macroeconometric modeling that took place in the 1970s (and its subsequent failure)
c) Economists actually believe that it is better to compete amongst themselves to build better models than to cooperate to build better models.
Friday, May 21, 2010
What I've always wondered about convergence
But was afraid to ask until it was asked for me: Why does anyone care about the distinction between convergence in probability and almost sure convergence?
Some answers:
1. "Suppose a person takes a bow and starts shooting arrows at a target. Let Xn be his score in n-th shot. Initially he will be very likely to score zeros, but as the time goes and his archery skill increases, he will become more and more likely to hit the bullseye and score 10 points. After the years of practice the probability that he hit anything but 10 will be getting increasingly smaller and smaller. Thus, the sequence Xn converges in probability to X = 10.Note that Xn does not converge almost surely however. No matter how professional the archer becomes, there will always be a small probability of making an error. Thus the sequence {Xn} will never turn stationary: there will always be non-perfect scores in it, even if they are becoming increasingly less frequent."
Also, almost sure convergence implies convergence in probability.
2. The most useful intuitive understanding I've been taught is that almost sure convergence guarantees that X_n be far from X (ie. further than any epsilon) only a finite number of times. Convergence in probability leaves open the possibility that X_n will be far from X an infinite number of times.
The best example I have to illustrate that is if you take Y_n as a Bernoulli(1/n) random variable. Clearly Y_n converges to 0 in probability, but it doesn't converge almost surely. Y_n will always be 1 for an infinite number of n's. You can see this from the second Borel-Cantelli Lemma.
Of course, I've got no idea if the distinction has any practical relevance for econometrics.
3. Convergence in probability is a form of weak convergence. Your students should understand the difference between convergence and weak convergence -- the difference is huge. If you have a sequence x_n, then weak convergence means that f(x_n) --> L for some f. This does not mean that x_n converges, but only that some attribute converges.
For example, you can ask, given N asset prices, if the sum of these prices converges to 1, does that mean that each individual asset price converges to something? No. Here f is the operation of taking the sum. It could be average, variance, integration against a test function, the infimum of a large set of integrations against test functions, whatever. ... Weak convergence, point-wise convergence, and uniform convergence are different concepts and useful ideas to understand, and they appear over and over again in different forms whatever branch of math you are studying.
Some answers:
1. "Suppose a person takes a bow and starts shooting arrows at a target. Let Xn be his score in n-th shot. Initially he will be very likely to score zeros, but as the time goes and his archery skill increases, he will become more and more likely to hit the bullseye and score 10 points. After the years of practice the probability that he hit anything but 10 will be getting increasingly smaller and smaller. Thus, the sequence Xn converges in probability to X = 10.Note that Xn does not converge almost surely however. No matter how professional the archer becomes, there will always be a small probability of making an error. Thus the sequence {Xn} will never turn stationary: there will always be non-perfect scores in it, even if they are becoming increasingly less frequent."
Also, almost sure convergence implies convergence in probability.
2. The most useful intuitive understanding I've been taught is that almost sure convergence guarantees that X_n be far from X (ie. further than any epsilon) only a finite number of times. Convergence in probability leaves open the possibility that X_n will be far from X an infinite number of times.
The best example I have to illustrate that is if you take Y_n as a Bernoulli(1/n) random variable. Clearly Y_n converges to 0 in probability, but it doesn't converge almost surely. Y_n will always be 1 for an infinite number of n's. You can see this from the second Borel-Cantelli Lemma.
Of course, I've got no idea if the distinction has any practical relevance for econometrics.
3. Convergence in probability is a form of weak convergence. Your students should understand the difference between convergence and weak convergence -- the difference is huge. If you have a sequence x_n, then weak convergence means that f(x_n) --> L for some f. This does not mean that x_n converges, but only that some attribute converges.
For example, you can ask, given N asset prices, if the sum of these prices converges to 1, does that mean that each individual asset price converges to something? No. Here f is the operation of taking the sum. It could be average, variance, integration against a test function, the infimum of a large set of integrations against test functions, whatever. ... Weak convergence, point-wise convergence, and uniform convergence are different concepts and useful ideas to understand, and they appear over and over again in different forms whatever branch of math you are studying.
Subscribe to:
Posts (Atom)