Beat the Press

Beat the press por Dean Baker

Beat the Press is Dean Baker's commentary on economic reporting. He is a Senior Economist at the Center for Economic and Policy Research (CEPR). To never miss a post, subscribe to a weekly email roundup of Beat the Press. Please also consider supporting the blog on Patreon.

I have a policy of not discussing items that directly refer to me in this blog, but I will make an exception today because the issues raised by Robert Samuelson are important. In his column Samuelson makes two key arguments. First, that the Reinhart-Rogoff conclusions about high debt leading to slow growth still stand even after the errors in the original paper were corrected, and second, that this work was never really the basis for austerity anyhow. Taking these in order, Samuelson constructs a chart showing the originally reported and corrected relationships between debt levels and GDP growth. Debt/GDP Annual economic growth, 1945-2009 ? Reinhart/Rogoff UMass economists 0-30% 4.1% 4.2% 30-60 2.8 3.1 60-90 2.8 3.2 90+ -0.1 2.2 He then tells readers: "After recalculating the Reinhart/Rogoff data, the UMass economists confirm that high debt implies lower economic growth." No, that is not right. The recalculated numbers show that high debt levels in the countries examined by Reinhart and Rogoff were associated with lower growth. However as the paper by Thomas Herndon, Michael Ash and Robert Pollin that exposed the error clearly explained, the growth falloff for countries with debt-to-GDP ratios above 90 percent was not statistically significant. In fact, they found a much stronger negative relationship between debt and GDP growth at very low ratios of debt-to-GDP. This means that if someone was basing policy on the corrected Reinhart-Rogoff numbers they would be arguing for debt-to-GDP levels in the range of 15-20 percent. That is not what Reinhart and Rogoff or anyone what else in this debate is saying. 
I have a policy of not discussing items that directly refer to me in this blog, but I will make an exception today because the issues raised by Robert Samuelson are important. In his column Samuelson makes two key arguments. First, that the Reinhart-Rogoff conclusions about high debt leading to slow growth still stand even after the errors in the original paper were corrected, and second, that this work was never really the basis for austerity anyhow. Taking these in order, Samuelson constructs a chart showing the originally reported and corrected relationships between debt levels and GDP growth. Debt/GDP Annual economic growth, 1945-2009 ? Reinhart/Rogoff UMass economists 0-30% 4.1% 4.2% 30-60 2.8 3.1 60-90 2.8 3.2 90+ -0.1 2.2 He then tells readers: "After recalculating the Reinhart/Rogoff data, the UMass economists confirm that high debt implies lower economic growth." No, that is not right. The recalculated numbers show that high debt levels in the countries examined by Reinhart and Rogoff were associated with lower growth. However as the paper by Thomas Herndon, Michael Ash and Robert Pollin that exposed the error clearly explained, the growth falloff for countries with debt-to-GDP ratios above 90 percent was not statistically significant. In fact, they found a much stronger negative relationship between debt and GDP growth at very low ratios of debt-to-GDP. This means that if someone was basing policy on the corrected Reinhart-Rogoff numbers they would be arguing for debt-to-GDP levels in the range of 15-20 percent. That is not what Reinhart and Rogoff or anyone what else in this debate is saying. 

I have had plenty of occasions to criticize the editorial views of both the NYT and the Washington Post, but there are major differences between these two pillars of the national media. It would be hard to find a better statement of these differences than the Times and Post‘s view of the impact of the sequester on air travel.

One could also add that the sequester is throwing around 600,000 people out of work according to the Congressional Budget Office. These are people who have the necessary skills to fill jobs in the economy but who will not be working because people in Washington lack the skills to design policies to keep the economy near full employment.

I have had plenty of occasions to criticize the editorial views of both the NYT and the Washington Post, but there are major differences between these two pillars of the national media. It would be hard to find a better statement of these differences than the Times and Post‘s view of the impact of the sequester on air travel.

One could also add that the sequester is throwing around 600,000 people out of work according to the Congressional Budget Office. These are people who have the necessary skills to fill jobs in the economy but who will not be working because people in Washington lack the skills to design policies to keep the economy near full employment.

One of the big issues left to be resolved in the debate over housing finance is the size of the down payment that homebuyers must put down in order for a mortgage to be considered a “qualified” mortgage. If a mortgage fits this definition, securitizers would not be required to hold capital against the mortgage.

NYT’s Dealbook had a post discussing the topic which highlighted research by Roberto G. Quercia, director of the Center for Community Capital at the University of North Carolina at Chapel Hill, which purports to show that there is not much additional risk of default with lower down payments. In fact, the numbers presented in the piece imply that Quercia’s research implies that low down payment loans have far higher risks of default. This means that lenders would either have to charge considerably higher interest rates or be subsidized for making these loans.

The NYT piece reports that Quercia found that the default rate during the years of the housing crash on a set of loans that met certain quality standards was 5.8 percent. However the default rate on loans with down payments of 20 percent or more was just 3.9 percent.

The piece reports that this higher down payment group comprised less than half of the loans in the study. If we assume that the 20 percent down payment group comprised 40 percent of the loans then this means that the default rate for home buyers putting less than 20 percent down was over 7.0 percent, more than 80 percent higher than for the 20 percent down payment group.

Furthermore, the losses for the less than 20 percent down payment group would be considerably higher on each default since there is less of a down payment cushion. If the loss for a default on a 20 percent down payment averages 25 percent, and we assume that the average down payment for the less than 20 percent group is 10 percent, then the losses for this group would average 39 percent of the amount loaned. (A 20 percent down payment is 25 percent of mortgage loan. A 10 percent down payment is just 11 percent of the amount loaned.)

Multiplying the increased probability of default (1.8) by the higher loss per default (1.6), the cost of default for the less than 20 percent down payment mortgages would be over 180 percent higher for the more than 20 percent down payment group, according to Quercia’s research. That would not seem to be a good argument to apply the same lending standards for this low down payment group.

It is also worth noting that Quercia’s assessment, as presented in the NYT post, that the 20 percent down payment would have excluded more than half of the borrowers does not follow from the evidence presented. If homebuyers knew that they had to put down 20 percent to get a lower cost mortgage, then they would be more likely to have a 20 percent down payment than in a context where this was not a requirement, as was true for the period examined by Quercia.  

One of the big issues left to be resolved in the debate over housing finance is the size of the down payment that homebuyers must put down in order for a mortgage to be considered a “qualified” mortgage. If a mortgage fits this definition, securitizers would not be required to hold capital against the mortgage.

NYT’s Dealbook had a post discussing the topic which highlighted research by Roberto G. Quercia, director of the Center for Community Capital at the University of North Carolina at Chapel Hill, which purports to show that there is not much additional risk of default with lower down payments. In fact, the numbers presented in the piece imply that Quercia’s research implies that low down payment loans have far higher risks of default. This means that lenders would either have to charge considerably higher interest rates or be subsidized for making these loans.

The NYT piece reports that Quercia found that the default rate during the years of the housing crash on a set of loans that met certain quality standards was 5.8 percent. However the default rate on loans with down payments of 20 percent or more was just 3.9 percent.

The piece reports that this higher down payment group comprised less than half of the loans in the study. If we assume that the 20 percent down payment group comprised 40 percent of the loans then this means that the default rate for home buyers putting less than 20 percent down was over 7.0 percent, more than 80 percent higher than for the 20 percent down payment group.

Furthermore, the losses for the less than 20 percent down payment group would be considerably higher on each default since there is less of a down payment cushion. If the loss for a default on a 20 percent down payment averages 25 percent, and we assume that the average down payment for the less than 20 percent group is 10 percent, then the losses for this group would average 39 percent of the amount loaned. (A 20 percent down payment is 25 percent of mortgage loan. A 10 percent down payment is just 11 percent of the amount loaned.)

Multiplying the increased probability of default (1.8) by the higher loss per default (1.6), the cost of default for the less than 20 percent down payment mortgages would be over 180 percent higher for the more than 20 percent down payment group, according to Quercia’s research. That would not seem to be a good argument to apply the same lending standards for this low down payment group.

It is also worth noting that Quercia’s assessment, as presented in the NYT post, that the 20 percent down payment would have excluded more than half of the borrowers does not follow from the evidence presented. If homebuyers knew that they had to put down 20 percent to get a lower cost mortgage, then they would be more likely to have a 20 percent down payment than in a context where this was not a requirement, as was true for the period examined by Quercia.  

The United States emits more than twice as much greenhouse gas per person as the average for West Europe. If countries like Germany, France, and Spain emitted as much per person as the United States, there would be no point worrying about greenhouse gas emissions because the planet would be fried. The world’s emissions levels over the last three decades would be so much larger that there would be no hope of reducing emissions enough to prevent catastrophic levels of global warming.

This fact (i.e. undeniably true statement) makes the Washington Post’s lecture to Europeans about the United States being “the world leader in fighting climate change” laughable. Apparently the Washington Post thinks the United States should get big points for improvement and somehow that Europe should emulate us even if we still are emitting far more greenhouse gas per person than Europe.

No one ever accused the Washington Post of having much grasp of logic and arithmetic. After all, they are doubling down in their support of austerity even after the Reinhart & Rogoff errors had been exposed. But this one seems to be below even the Post’s incredibly low standard. Come on, we know that the Post hates Europe because they don’t give all of their money to the rich, and ordinary workers can expect to get health care and a decent retirement, but this one is really off the deep end.

The United States emits more than twice as much greenhouse gas per person as the average for West Europe. If countries like Germany, France, and Spain emitted as much per person as the United States, there would be no point worrying about greenhouse gas emissions because the planet would be fried. The world’s emissions levels over the last three decades would be so much larger that there would be no hope of reducing emissions enough to prevent catastrophic levels of global warming.

This fact (i.e. undeniably true statement) makes the Washington Post’s lecture to Europeans about the United States being “the world leader in fighting climate change” laughable. Apparently the Washington Post thinks the United States should get big points for improvement and somehow that Europe should emulate us even if we still are emitting far more greenhouse gas per person than Europe.

No one ever accused the Washington Post of having much grasp of logic and arithmetic. After all, they are doubling down in their support of austerity even after the Reinhart & Rogoff errors had been exposed. But this one seems to be below even the Post’s incredibly low standard. Come on, we know that the Post hates Europe because they don’t give all of their money to the rich, and ordinary workers can expect to get health care and a decent retirement, but this one is really off the deep end.

The Bureau of Economic Analysis (BEA) will adopt a new methodology for measuring GDP this summer. The methodology will treat research and development and the creation of artistic works as forms of capital that depreciate through time rather than one-time expenditures. This will lead to an increase in measured GDP of close to 3.0 percent according to BEA's analysis. There are three points worth making on this change. First, for you conspiracy buffs, this one has been in the works for close to two decades. The government didn't just come up with it to make President Obama look better. Go back to digging up the Real Story about the plunge in gold prices. The second point is that the methodology for this will inevitably be very troubling. If Pfizer has a patent for a great new cancer drug we will now pick this up as an increase in the investment component of GDP. Suppose Merck develops a drug that does the exact same thing, except that it gets around Pfizer's patent. According to the new methodology this would further increase GDP. Of course, this is a battle over rents, not actually an increase in total output. That is a problem. Expenditures for rent-seeking don't make us richer in aggregate. In fact, this is already a problem now, it's just likely to be more of a problem in the future. Consider the situation where a software developer makes their great new software available for free. Our friends over at BEA won't show any gain to GDP even though our living standards will certainly be improved by much more than if they had patented the software and charged for it.
The Bureau of Economic Analysis (BEA) will adopt a new methodology for measuring GDP this summer. The methodology will treat research and development and the creation of artistic works as forms of capital that depreciate through time rather than one-time expenditures. This will lead to an increase in measured GDP of close to 3.0 percent according to BEA's analysis. There are three points worth making on this change. First, for you conspiracy buffs, this one has been in the works for close to two decades. The government didn't just come up with it to make President Obama look better. Go back to digging up the Real Story about the plunge in gold prices. The second point is that the methodology for this will inevitably be very troubling. If Pfizer has a patent for a great new cancer drug we will now pick this up as an increase in the investment component of GDP. Suppose Merck develops a drug that does the exact same thing, except that it gets around Pfizer's patent. According to the new methodology this would further increase GDP. Of course, this is a battle over rents, not actually an increase in total output. That is a problem. Expenditures for rent-seeking don't make us richer in aggregate. In fact, this is already a problem now, it's just likely to be more of a problem in the future. Consider the situation where a software developer makes their great new software available for free. Our friends over at BEA won't show any gain to GDP even though our living standards will certainly be improved by much more than if they had patented the software and charged for it.
That is quite literally what he told us in his column. His second paragraph tells readers: "Among economists, there is no consensus on policies. Is “austerity” (government spending cuts and tax increases) self-defeating or the unavoidable response to high budget deficits and debt? Can central banks such as the Federal Reserve or the European Central Bank engineer recovery by holding short-term interest rates near zero and by buying massive amounts of bonds (so-called “quantitative easing”)? Or will these policies foster financial speculation, instability and inflation? The public is confused, because economists are divided." See, we don't know what to do, so we just can't do anything. All those suckers who are unemployed or seeing stagnant wages, well we just don't know. And the fact that those on the top are getting rich with 60-year high shares of national income, well what can we do about that? It's just too confusing. While Samuelson may be very confused by economics, those who understood their intro econ have little difficulty explaining the current situation. The housing bubble was driving the economy prior to its collapse. The collapse eliminated more than $600 billion in demand from residential construction and more than $500 billion in demand from consumption. There was also demand lost from a collapse of a smaller bubble in non-residential construction and from state and local government cutbacks forced by a loss of tax revenue. This is not complicated and it was predicted.
That is quite literally what he told us in his column. His second paragraph tells readers: "Among economists, there is no consensus on policies. Is “austerity” (government spending cuts and tax increases) self-defeating or the unavoidable response to high budget deficits and debt? Can central banks such as the Federal Reserve or the European Central Bank engineer recovery by holding short-term interest rates near zero and by buying massive amounts of bonds (so-called “quantitative easing”)? Or will these policies foster financial speculation, instability and inflation? The public is confused, because economists are divided." See, we don't know what to do, so we just can't do anything. All those suckers who are unemployed or seeing stagnant wages, well we just don't know. And the fact that those on the top are getting rich with 60-year high shares of national income, well what can we do about that? It's just too confusing. While Samuelson may be very confused by economics, those who understood their intro econ have little difficulty explaining the current situation. The housing bubble was driving the economy prior to its collapse. The collapse eliminated more than $600 billion in demand from residential construction and more than $500 billion in demand from consumption. There was also demand lost from a collapse of a smaller bubble in non-residential construction and from state and local government cutbacks forced by a loss of tax revenue. This is not complicated and it was predicted.

The NYT told readers about efforts to require city governments to disclose their pension liabilities in order to be able to continue to issue tax-exempt bonds. While better disclosure of pension liabilities may be desirable, there are important methodological issues that the piece neglects to mention.

The piece notes that Moody’s, the bond-rating agency, is now adjusting pension liabilities reported by city governments in making assessments of their financial situation. The methodology used by Moody’s to make its adjustment ignores the expected return from market assets. Moody’s methodology would have implied that pension funds were hugely over-funded at the peak of the stock bubble in 2000, even though any reasonable assessment of pension liabilities would have predicted very low returns on assets at that point.

If municipalities were to adjust pension contributions to sustain funding targets using the Moody’s methodology it would lead to highly erratic funding patterns. In most cases, governments would have to make large contributions for a number of years and then would have to contribute little or nothing as market returns exceeded the return assumptions used by Moody’s. This would be comparable to having city governments accumulate large bank deposits so that they could pay for their schools or fire departments from the annual interest.

Usually public finance economists advocate funding mechanisms that imply an even flow through time so that no particular cohort of taxpayers is excessively benefited or penalized. The Moody’s methodology, which is also being pushed as part of the bill being debated in Congress, goes in the opposite direction. This fact should have been noted in the piece.

The NYT told readers about efforts to require city governments to disclose their pension liabilities in order to be able to continue to issue tax-exempt bonds. While better disclosure of pension liabilities may be desirable, there are important methodological issues that the piece neglects to mention.

The piece notes that Moody’s, the bond-rating agency, is now adjusting pension liabilities reported by city governments in making assessments of their financial situation. The methodology used by Moody’s to make its adjustment ignores the expected return from market assets. Moody’s methodology would have implied that pension funds were hugely over-funded at the peak of the stock bubble in 2000, even though any reasonable assessment of pension liabilities would have predicted very low returns on assets at that point.

If municipalities were to adjust pension contributions to sustain funding targets using the Moody’s methodology it would lead to highly erratic funding patterns. In most cases, governments would have to make large contributions for a number of years and then would have to contribute little or nothing as market returns exceeded the return assumptions used by Moody’s. This would be comparable to having city governments accumulate large bank deposits so that they could pay for their schools or fire departments from the annual interest.

Usually public finance economists advocate funding mechanisms that imply an even flow through time so that no particular cohort of taxpayers is excessively benefited or penalized. The Moody’s methodology, which is also being pushed as part of the bill being debated in Congress, goes in the opposite direction. This fact should have been noted in the piece.

Ilyana Kuziemko and Stefanie Stantcheva seem determined to win the contest for greatest out to lunch op-ed column of all time. They note the huge rise in inequality in wealth and income over the last three decades. They tell readers that the public is aware of this increase. However they say that the public doesn’t trust the government to do anything to address inequality:

“Whether or not the rise in inequality has itself lowered Americans’ faith in government, the low opinion in which Americans hold their government may well limit their willingness to connect concern with inequality to government action.”

Sorry folks, but the public is way ahead of our researchers here. The public recognizes that inequality did not just happen, as the our oped authors seem to believe. Inequality was the result of deliberate government policies. For example, we have high unemployment now because of government policies that are reducing demand in the economy. Manufacturing workers lost jobs and have lower wages because of trade policy designed to put them in competition with low-paid workers in the developing world while protecting our doctors and lawyers.

Maybe if social scientists who did surveys understood a bit more economics, they would ask better questions.

Ilyana Kuziemko and Stefanie Stantcheva seem determined to win the contest for greatest out to lunch op-ed column of all time. They note the huge rise in inequality in wealth and income over the last three decades. They tell readers that the public is aware of this increase. However they say that the public doesn’t trust the government to do anything to address inequality:

“Whether or not the rise in inequality has itself lowered Americans’ faith in government, the low opinion in which Americans hold their government may well limit their willingness to connect concern with inequality to government action.”

Sorry folks, but the public is way ahead of our researchers here. The public recognizes that inequality did not just happen, as the our oped authors seem to believe. Inequality was the result of deliberate government policies. For example, we have high unemployment now because of government policies that are reducing demand in the economy. Manufacturing workers lost jobs and have lower wages because of trade policy designed to put them in competition with low-paid workers in the developing world while protecting our doctors and lawyers.

Maybe if social scientists who did surveys understood a bit more economics, they would ask better questions.

The NYT has difficulty finding pundits who can write knowledgeably about economics. Thomas Friedman made this point in his Sunday column. At one point he quotes Gary Green, the president of Forsyth Technical Community College, in Winston-Salem, N.C.: "'We have a labor surplus in this country and a labor shortage at the same time,' Green explained to me. Workers in North Carolina, particularly in textiles and furniture, who lost jobs either to outsourcing or the recession in 2008, often 'do not have the skills required to get a new job today' in the biotech, health care and manufacturing centers that are opening in the state. "If before, he added, 'you just needed a high school shop class or a short postsecondary certificate to work in a factory, now you need an associate degree in machining,' a two-year program that requires higher math, I.T. and systems skills. In addition, some employers are now demanding that you not only have an associate degree but that nationally recognized skill certifications be incorporated into the curriculum to show that you have mastered the skills they want, like computer-integrated machining." Actually there are simple ways to identify labor shortages. First and foremost we should be seeing rapidly rising wages. If employers cannot get the workers they need then they raise the wages they offer to pull workers away from other employers. This is how markets work. (We should also see longer workweeks and increased vacancies.) In fact there is no major sector of the economy where wages are rising rapidly. This shows rather conclusively that workers do not have skill shortages although it may be the case that many managers are so ignorant of markets that they don't know that the way to attract better workers is to raise wages. Of course that would suggest the need to better train managers, not workers.
The NYT has difficulty finding pundits who can write knowledgeably about economics. Thomas Friedman made this point in his Sunday column. At one point he quotes Gary Green, the president of Forsyth Technical Community College, in Winston-Salem, N.C.: "'We have a labor surplus in this country and a labor shortage at the same time,' Green explained to me. Workers in North Carolina, particularly in textiles and furniture, who lost jobs either to outsourcing or the recession in 2008, often 'do not have the skills required to get a new job today' in the biotech, health care and manufacturing centers that are opening in the state. "If before, he added, 'you just needed a high school shop class or a short postsecondary certificate to work in a factory, now you need an associate degree in machining,' a two-year program that requires higher math, I.T. and systems skills. In addition, some employers are now demanding that you not only have an associate degree but that nationally recognized skill certifications be incorporated into the curriculum to show that you have mastered the skills they want, like computer-integrated machining." Actually there are simple ways to identify labor shortages. First and foremost we should be seeing rapidly rising wages. If employers cannot get the workers they need then they raise the wages they offer to pull workers away from other employers. This is how markets work. (We should also see longer workweeks and increased vacancies.) In fact there is no major sector of the economy where wages are rising rapidly. This shows rather conclusively that workers do not have skill shortages although it may be the case that many managers are so ignorant of markets that they don't know that the way to attract better workers is to raise wages. Of course that would suggest the need to better train managers, not workers.
The NYT appears to be following the pattern of journalism practiced by the Washington Post in openly editorializing in its news section. Today the news section features a diatribe against the Danish welfare state that is headlined, "Danes Rethink a Welfare State Ample to a Fault." There's not much ambiguity in that one. The piece then proceeds to present a state of statistics that are grossly misleading and excluding other data points that are highly relevant. The first paragraphs describe the generosity of the welfare state, then we get this ominous warning in the 5th paragraph: "But Denmark’s long-term outlook is troubling. The population is aging, and in many regions of the country people without jobs now outnumber those with them." Oooooh, scary! Yeah people are living longer in Denmark, that's something that's been happening for a couple of hundred years or so. Like every other wealthy country, people live longer in Denmark than in the United States. While they are projected to continue to see gains in life expectancy and further aging of the population, the increase is actually going to much slower than in the United States.
The NYT appears to be following the pattern of journalism practiced by the Washington Post in openly editorializing in its news section. Today the news section features a diatribe against the Danish welfare state that is headlined, "Danes Rethink a Welfare State Ample to a Fault." There's not much ambiguity in that one. The piece then proceeds to present a state of statistics that are grossly misleading and excluding other data points that are highly relevant. The first paragraphs describe the generosity of the welfare state, then we get this ominous warning in the 5th paragraph: "But Denmark’s long-term outlook is troubling. The population is aging, and in many regions of the country people without jobs now outnumber those with them." Oooooh, scary! Yeah people are living longer in Denmark, that's something that's been happening for a couple of hundred years or so. Like every other wealthy country, people live longer in Denmark than in the United States. While they are projected to continue to see gains in life expectancy and further aging of the population, the increase is actually going to much slower than in the United States.

Want to search in the archives?

¿Quieres buscar en los archivos?

Click Here Haga clic aquí