Saturday, July 30, 2016

E.g., Abandon i.e. and Other Latinisms, etc.

I feel so vindicated.

Someone named Persis Howe recently blogged at a UK government website that the style-guide for the website is being updated to recommend avoiding the use of e.g., i.e., and etc. For the last 30 years I've been editing articles at the Journal of Economic Perspectives, I've been discouraging writers from using these terms, too. Howe notes that a growing number of people are having content read to them by audio programs, which often mangle these terms. She also writes:
We promote the use of plain English on GOV.UK. We advocate simple, clear language. Terms like eg, ie and etc, while common, make reading difficult for some. Anyone who didn’t grow up speaking English may not be familiar with them. Even those with high literacy levels can be thrown if they are reading under stress or are in a hurry - like a lot of people are on the web.
Of course, now that my desire to purge plain English of Latinisms has won this small victory, my horizons are expanding. A number of economists apparently have a psychological need to write "ex ante" and "ex post," rather than using the vocabulary of commonplace words that describe sequences in either temporal or expectational terms, including "beforehand" or "expected" or "before the fact," as well as "afterwards" or "realized" or "occurred later."  And yes, these issues actually raise the blood pressure of pedants like me.

Friday, July 29, 2016

Public Pensions on Shaky Ground

The stock market run-up of the 1990s was fool's gold for many state and local pension funds. At the height of the dot-com boom, the typical pension fund had enough on hand to cover all of its expected future costs. But booms don't tend to last, and that one didn't, either. There are a couple of short recent reports that offer a useful update on the current status of public pension funds. One is the "Issue Brief" by Alicia H. Munnell and Jean-Pierre Aubry called "The Funding of State and LocalPensions: 2015-2020," published by the  Center for State & Local Government Excellence in June 2016. The other, by William G. Gale and Aaron Krupkin, is called "Financing State and Local Pension Obligations:Issues and Options," and was published as a Brookings Institution Working Paper in July 2016.

The basic calculation here is to look at the assets that public pension funds have on hand, and to compare that amount with what is needed to pay the benefits that have already been promised to be paid in the future. To make this calculation, you need to make a guess about what rate of return will be earned by the assets currently on hand. A typical current estimate is that pension fund assets will earn an average of 7.6% per year for decades to come. We'll return to the realism of that number in a moment. But using it as a given, Munnell and Aubry present calculations that the average state and local pension fund has on hand about 74% of what it needs to pay the benefits that have already been promised. (The exact percentage varies a bit because of some alternative accounting rules.)

Notice that when the stock market peaked right around 2000, there was a golden moment when public pension funds were fully funded. But rather than build on that moment, by assuring that the funds would remain fully funded into the future, a number of state and local governments saw this as a chance to promise higher pension benefits and to make lower contributions to pension funds. Apparently these actions were acceptable both to elected officials and to the leadership of public employee unions. And now here we are.

Of course, this overall average is mixture of some funds that are doing better, and some doing worse. For example, here are examples of public pension funds that are less than 50% funded in 2015: Arizona Public Safety Personnel (49% funded), Chicago Municipal Employees (37%), Chicago Police (27%), Connecticut State Employees Retirement system (43%), Illinois State Employees Retirement System (36%), Kentucky Employee Retirement System (22%), and the Philadelphia Municipal Retirement System (44%).

And what if the pension fund assets on hand don't earn 7.6% per year? As Munnell and Aubry write: "Public pensions currently hold about 70 percent of their assets in risky investments, including more than half of their assets in equities. As discussed, on average, plans assume a nominal return of 7.6 percent on their whole portfolios, which implies nominal stock returns of 9.6 percent. In contrast, many investment firms project much lower equity returns ..."

By their calculations, if one assumes only a 6% annual rate of return on pension fund assets going forward, then the average funds are only 58% funded at present. And if one assumes that the rate of return on pension fund assets is only 4% per year going forward, then the average fund is only 45% funded at present. Of course, those pension funds with below-average funding at present would be even worse off if returns are lower than hoped-for. As Munnell and Aubry write: "What happens from here on out depends very much on investment performance."

Assuming that we will not have the pleasant surprise of very high investment returns that rescue the pension funds, what policy options are possible? Gale and Krupkin walk through the choices. As they write: "The implications of the projections are unpleasant but straightforward. Governments that face significant shortfalls will have to cut employee benefits, raise employee contributions, or finance higher employer contributions with tax increases or spending cuts. The eventual changes may be modified (or hidden) by pension reform, but the basic direction of the required changes is clear."
Ultimately, the argument is over who will bear the shortfall in public pension funds. The candidates are current retirees, current government workers, previous government workers who haven't yet claimed benefits, or the public through either taxes that are higher or government services that are lower than they would otherwise be.

I don't have a good sense of what sort of deals should be cut when some public pension funds inevitably can't fulfill their promises. But I do know that a hefty dose of the blame should go to the specific state or local coalitions of elected officials, public sector unions, and voters that found it easy to promise future payments, but apparently impossible to assure that sufficient funds were put aside for those promises. In many places around the country, public pension funds have been run with a reasonably high degree of responsibility (if not always quite as prudently as I would have preferred if I was a state employee), or where meaningful reforms in the direction of pension solvency have already been undertaken.  But where the behavior has been imprudent, those state and local governments should have to face their own voters, and those public unions should have to face their own workers.

Thursday, July 28, 2016

Adam Smith on Human Capacity for Self-Deceit

Adam Smith offered a characteristically pungent insight on the subject of the human capacity for self-deceit in his 1759 book, The Theory of Moral Sentiments. I quote here from the always-useful 1790 edition available online at the Library of Economics and Liberty website. Here's Smith in TMS  (1759 [1790], part III, Ch. 1).
"The opinion which we entertain of our own character depends entirely on our judgments concerning our past conduct. It is so disagreeable to think ill of ourselves, that we often purposely turn away our view from those circumstances which might render that judgment unfavourable. He is a bold surgeon, they say, whose hand does not tremble when he performs an operation upon his own person; and he is often equally bold who does not hesitate to pull off the mysterious veil of self-delusion, which covers from his view the deformities of his own conduct. Rather than see our own behaviour under so disagreeable an aspect, we too often, foolishly and weakly, endeavour to exasperate anew those unjust passions which had formerly misled us; we endeavour by artifice to awaken our old hatreds, and irritate afresh our almost forgotten resentments: we even exert ourselves for this miserable purpose, and thus persevere in injustice, merely because we once were unjust, and because we are ashamed and afraid to see that we were so. … 
"This self-deceit, this fatal weakness of mankind, is the source of half the disorders of human life. If we saw ourselves in the light in which others see us, or in which they would see us if they knew all, a reformation would generally be unavoidable. We could not otherwise endure the sight."
It may be that modest degree of self-deceit about our own capabilities and appearance helps many of us to get up in the morning and face the day. But self-deceptions have an unpleasant habit of colliding with reality, sooner or later. One hopes that those collisions with reality can be gentle, and that they can be an opportunity for what Smith called "a reformation" and what we now label as a "personal growth opportunity." But it's easy to think of situations where people become so highly invested in self-deception about their own conduct that, when the collision with reality occurs, they push back with anger and counter-accusations and retreat further into their self-deceit, rather than engaging in self-examination. Indeed, I offer as a hypothesis that 21st-century culture may in various ways encourage self-deceit over self-examination. 

Wednesday, July 27, 2016

International Trade as a Scapegoat

The ferocity of some of the arguments in the US over global trade can be a little surprising to me. After all, the US economy with its enormous internal market is considerably less exposed to international trade than the world average. Here's a figure generated from the World Bank website showing imports/GDP for the world economy as a whole, and the US economy.  For the world economy on average, the import/GDP ratio is approaching 30%; for the US economy, the import/GDP ratio is higher than it ways, but still half the global level.

Moreover, in a Gallup poll earlier this year, a strong majority of Americans are more likely to perceive foreign trade as an opportunity for growth than as a threat. Moreover, the pro-trade majority has been rising since 2008. I do need to add in passing that pretty much all economists would view the specific Gallup question, which assumes that exports benefit the US economy and imports threaten it, as a fundamentally wrong-headed view of why an economy benefits from trade. But there are similar pro-trade majorities in other polls, like this recent NBC News/Wall Street Journal poll,

Doug Irwin offers a useful overview of the pro-trade position in his essay in the July/August 2016 issue of Foreign Affairs, "The Truth About Trade: What Critics Get Wrong About the Global Economy." He writes: "By and large, the United States has no major difficulties with respect to trade, nor does it suffer from problems that could be solved by trade barriers. What it does face, however, is a much larger problem, one that lies at the root of anxieties over trade: the economic ladder that allowed previous generations of lower-skilled Americans to reach the middle class is broken."

My usual way of making this point is to argue that international trade, and in particular arguments over the details of trade agreements, is an easy scapegoat for more profound and harder-to-tackle economic dislocations. I'd be delighted if America's economic issues could be resolved by, say, renegotiating or just not signing some trade agreement. But I don't believe. it.

Irwin works through many of the concerns raised about international trade, like those who mistakenly believe that the US trade deficit was a main cause of job loss during the Great Recession. He writes:
In fact, the trade deficit usually increases when the economy is growing and creating jobs and decreases when it is contracting and losing jobs. The U.S. current account deficit shrank from 5.8 percent of GDP in 2006 to 2.7 percent in 2009, but that didn’t stop the economy from hemorrhaging jobs. And if there is any doubt that a current account surplus is no economic panacea, one need only look at Japan, which has endured three decades of economic stagnation despite running consistent current account surpluses.
But to me, Irwin's key point is that the big underlying issue disrupting the US economy is technological change. Indeed, the changes in communication, logistics, transportation, and information processing are dramatically altering the US economy all by themselves. Indeed, developments in technology are a large part of what make global supply chains viable in the first place.International trade is part of the picture, too, but competition from robots and computers are ultimately a bigger disruptor than competition from workers in China or India.  Irwin writes:
"Although imports have put some people out of work, trade is far from the most important factor behind the loss of manufacturing jobs. The main culprit is technology. Auto­mation and other technologies have enabled vast productivity and efficiency improvements, but they have also made many blue-collar jobs obsolete. One representative study, by the Center for Business and Economic Research at Ball State University, found that pro­ductivity growth accounted for more than 85 percent of the job loss in manufacturing between 2000 and 2010, a period when employment in that sector fell by 5.6 million. Just 13 percent of the overall job loss resulted from trade, although in two sectors, apparel and furniture, it accounted for 40 percent.
"Although the United States boasts a highly skilled work force and a solid technological base, it is still the case that only one in three American adults has a college education. In past decades, the two-thirds of Americans with no postsecondary degree often found work in manufacturing, construction, or the armed forces. ... Over time, however, these opportunities have disappeared. Technology has shrunk manufacturing as a source of large-scale employ­ment: even though U.S. manufacturing output continues to grow, it does so with many fewer workers than in the past. Construction work has not recovered from the bursting of the housing bubble. And the military turns away 80 percent of applicants due to stringent fitness and intelligence requirements. There are no comparable sectors of the economy that can employ large numbers of high-school-educated workers.
"This is a deep problem for American society. The unemployment rate for college-educated workers is 2.4 percent, but it is more than 7.4 percent for those without a high school diploma—and even higher when counting discouraged workers who have left the labor force but wish to work. These are the people who have been left behind in the twenty-first-century economy—again, not primarily because of trade but because of structural changes in the economy. Helping these workers and ensuring that the economy delivers benefits to everyone should rank as urgent priorities. But here is where the focus on trade is a diversion. Since trade is not the underlying problem in terms of job loss, neither is protectionism a solution."
I confess that I have my "stop the world, I want to get off" moments. But the rest of the world economy isn't going to stop. Whether or not the disruptive effects of technology continue to moves ahead in the US, technology is going to be developed and adopted elsewhere. If the US decides to take political actions to reduce its exposure to foreign trade, other countries are going to keep signing such agreements and building global supply chains. There are real challenges for how to create  ladders of opportunity for successful careers--not just jobs paid by the hour--for workers across the spectrum of education and skills. And there are legitimate policy disputes concerning the fine print of what's in various trade agreements. But backing away from technology and the global economy is not a successful path to prosperity.

Tuesday, July 26, 2016

High-Skilled Immigration

On one side, there seems to be near-universal agreement that the US economy would benefit from workers who had higher skill levels. But if the rising skill levels are generated by the  immigration of high-skilled workers, this consensus can become wobbly.  The National Academy of Sciences offers a useful overview of these issues in Immigration Policy and the Search for Skilled Workers: Summary of a Workshop, published late in 2015. As the title implies, this report is a description of a conference, and most of the report is in the form of having the rapporteurs, Gail Cohen, Aqila Coulthurst, and Joe Alper, paraphrases presentations made at the conference.

High-skilled immigration is tied both to education and to the labor market: if a country like the United States welcomes foreign students to American colleges and universities, as undergraduates, graduate students, and faculty, there will inevitably be more situations where US-based companies want to hire this foreign-born but geographically available talent. Here are a couple of illustrative figures from an presentation by Lindsay Lowell. The left-hand panel shows that the US attracts by far the largest number of international students in total terms. The right-hand panel shows that when focusing just on science, technology, engineering, and mathematics students, the US is still near the top in the percentage of those students who are international.
One result of this influx of foreign talent is that the enormous US economy, shown by the red dotted line below, is among the world leaders in the share of its workforce who fall into the broad job category of "researchers"-- which is presumably a good thing in the coming knowledge economy.

Richard Freeman described this education-to-employment connection for technology-based skills in his presentation, paraphrased like this:

U.S. National Science Foundation estimates that 63 percent of all post-doctoral STEM students working in U.S. universities are international students, and that 49 percent of international post-doctoral fellows received their PhDs in the United States. There has been a corresponding increase in the number of scientific papers coming from U.S. laboratories that have Chinese co-authors or coauthors from other emerging economies. These international students are not merely getting an education in the United States—they are also becoming U.S. STEM workers after graduation. In 2005, over a third of all STEM workers with PhDs were foreign born, with 64 percent receiving their PhD from U.S. universities. Over a quarter of U.S. STEM workers with Master’s degrees were born in another country and 15 percent of foreign-born STEM workers with Master’s
degrees received that degree in the United States. According to a different dataset, the percentage of foreign-born workers in U.S. STEM jobs increased from 11 percent to 19 percent between 1990 and 2011 for those with Bachelor’s degrees, from 19 percent to 34.3 percent for those with Master’s degrees, and from 24 percent to 43 percent for PhDs.
Lowell noted that after STEM fields, business was the next most-popular field for high-skill immigrants. Here's a paraphrase: "After STEM fields, business was the most popular subject of study for international students in the United States during that period. The impact of a large number of business students may be substantial on growth, because it is often the business majors who take advantage of ideas and bring them to market ..."

The economics of immigration involves evaluating a set of tradeoffs. Do immigrants help the economy to grow, for example by allowing native workers in the economy to specialize in ways that potentially raise productivity and wages for everyone? Or do immigrants only compete for existing jobs in a way that reduces job prospects and standard of living for native workers? As William Kerr points out in his presentation, there are different historical examples of each of these. A study of the chemists who fled Nazi Germany for the United States suggest that they helped the US chemical industry to grow substantially. A study of the wave of Russian mathematicians who came to the US in the 1990s suggests wages and job opportunities for native-born US mathematicians were reduced as a result.

When looking only at high-skill immigration, it seems clearly beneficial to an economy to have immigrants who are also gifted entrepreneurs, building companies that provide jobs and secure high-wage employment. Moreover, there seem to be what economists call "agglomeration effects" in technology, where a group of people with interrelated technical skills all come together in one place, there can be an ongoing growth of innovation and production that exceeds what this group would have accomplished if they were dispersed. To put this in concrete terms, it's a good thing for the US economy that the Silicon Valley agglomeration, which relies heavily on an influx of technical and business talent from all around the world, is located in this country.

The less clear-cut case involves what might be called undistinguished high-skill immigrants--that is, someone who is at best an average computer programmer or laboratory researcher. By definition, the undistinguished are less likely to create companies or be a key ingredient in an agglomeration. However, they may well compete with average native high-skill workers for jobs and wage. But here as well, the question is whether high-skilled immigrants may in some ways be complementary with high-skilled native labor.

A lot of the NAS report considers public policies from different countries about high-skilled immigration. The US stands out as a country that has not been especially encouraging to high-skilled immigration, but seems to get a disproportionate share of those immigrants nonetheless. As the report points out, in the United States, about 70% of immigration is family related, another 15% is humanitarian, and the remaining 15% is employment-based (which includes temporary high-skilled immigrants). In Canada and Australia, by contrast, about 30-40% of immigration is family-based or humanitarian, and the remaining 60-70% is employment-based. But as Lowell noted (according to this paraphrase), the US still does very well in the global contest for talent:
"Another indication of how well the United States is competing for international STEM workers comes from data on the number of high-skilled foreign-born workers in the 20 leading destination nations. From 1980 to 2010, the percentage of high-skilled migrants living in the United States relative to the other top destinations rose from 46 percent to 49 percent, even as the total number rose by more than four-fold. Similarly, data from the World Intellectual Property Organization showed that from 2001 to 2010, the flow of inventors around the world was dominated by flow into the United States, while OECD data shows that the United States remains the main destination for international
authors of scientific papers."
Pia Orrenius made the point that while the US immigration system for attracting high-skill immigrants is not especialy welcoming, the US makes up for it by being more welcoming to high-skill immigrants in other ways. Here's a paraphrase:
Immigration policy is just one tool of many that can result in a better, more qualified, nimble and innovative workforce. Luckily for the United States, the nation does well in other areas—the quality of our institutions of higher education, the salaries that U.S. employers pay, the flexible labor markets with many job opportunities, and the relative ease with which foreign workers integrate in the U.S. workforce, among others—that enable the country to be competitive in the international market for high-skilled workers.
In the past, policy arguments over high-skilled immigration have often been jumbled together with overall arguments about comprehensive immigration reform, but the issues raised are not the same. Higher education is expanding dramatically around the world, emerging-market economies are growing more rapidly than the world average, and global talent pool is expanding quickly, too. Competition for where these workers choose to locate will be real and ongoing. But in the 21st-century global economy, only some of these high-skill workers will not be planning to immigrate permanently. Many other will be seeking to make connections and build experience, and then moving elsewhere. In this sense, the policy issues of  high-skilled immigration are often not about permanent migration, but instead are about flexibility of work arrangements and geographic locations in an interconnected world.

At the NAS conference, Madeleine Sumption offered the intriguing thought that the US system of enticing high-skill immigrants through a mixture of educational and business opportunities, along with temporary work visas, may be the broadly the right approach for talent in the global economy. But in her view, the existing US approaches to high-skill migrants needs an overhaul with a big dose of additional flexibility. Sumption said: "The U.S. has the right model, it is just falling apart. ... We need to fix that model rather than think of something totally new.”

Thursday, July 21, 2016

An Update on Costs of End-of-Life Care

For those interested in the health care costs in end-of-life care, Medicare data are the obvious place to look.Of the 2.6 million people who died in the U.S. in 2014,  2.1 million, or eight out of 10, were people on Medicare, making Medicare the largest insurer of medical care provided at the end of life. Spending on Medicare beneficiaries in their last year of life accounts for about 25% of total Medicare spending on beneficiaries age 65 or older." Juliette Cubanski, Tricia Neuman, Shannon Griffin, and Anthony Damico make this point at the start of their short "data note" entitled "Medicare Spending at the End of Life: A Snapshot of Beneficiaries Who Died in 2014 and the Cost of Their Care" (July 2016, published by the Kaiser Family Foundation). "

Average health care spending for Medicare recipients who died in 2014 was $34,529, nearly four times as high as the average Medicare spending of $9,121 for those who didn't die. This general pattern isn't surprising: after all, those who die often tend to have health issues beforehand.  The detailed data shows that the biggest part of this cost difference is driven by higher spending for in-patient care in hospitals for those who died in 2014. What's interesting to me is that the share of Medicare spending going to those who die in that year seems to be diminishing.

Figure 3: The share of total traditional Medicare spending on traditional Medicare beneficiaries who died at some point in the year has declined over time

What explains this shift? The report lists these causes:
"In addition, we find that total spending on people who die in a given year accounts for a relatively small and declining share of traditional Medicare spending. This reduction is likely due to a combination of factors, including: growth in the number of traditional Medicare beneficiaries overall as the baby boom generation ages on to Medicare, which means a younger, healthier beneficiary population, on average; gains in life expectancy, which means beneficiaries are living longer and dying at older ages; lower average per capita spending on older decedents compared to younger decedents; slower growth in the rate of annual per capita spending for decedents than survivors, and a slight decline between 2000 and 2014 in the share of beneficiaries in traditional Medicare who died at some point in each year."
(A couple of notes here: 1) The graph and all the data here refer to "traditional Medicare," which is the two-thirds of Medicare recipients who are not in "Medicare Advantage" plans. In traditional Medicare, the government pays health care providers on a fee-for-service basis, and thus has good data on what the costs were for services each year. In Medicare Advantage, Medicare makes monthly insurance-like payments to a managed care organization--like a health maintenance organization--and so the government does not have readily available data on the costs of what actual health care was provided at any given time. 2) The 13.5% in the graph above for 2014 doesn't match the 25% at the top. The difference is that this figure looks at the health care costs incurred in 2014 for those who died in 2014. The 25% figure refers to health care costs incurred in the 12 months before death--which usually reaches back into the previous year. For looking at trends, either approach can work fine, but plotting data for costs in the 12 months before death and comparing it to other spending in the same time interval is a more complicated tas, and the official data is organized on an annual basis, so that's what is reported here.)  

A misconception which seems popular, at least based on the kind of questions I hear, is that end-of-life spending is especially high for the very elderly. That doesn't seem to be true. This figures shows spending of those who died in 2014 by age: for example, Medicare spending on 65 year-olds who died in 2014 averaged $38,840, while for those over age 100 it averaged $14,985. Conversely, Medicare spending on those who survived 2014 tends to rise by age.
Figure 9: Medicare per capita spending for decedents over age 65 declined with age in 2014, while spending for survivors increased
This pattern seems like a positive one to me, in the sense that I suspect there is more that health care can do for the average person who is 65 or 70, compared to the average person who is 100 or 105. A more detailed breakdown of this data shows that when just looking at health care costs of those who died in 2014 by age, those who were in their late 60s had much higher expenditures on in-patient hospital costs (the orange bars), while the older age groups tended to have higher spending on hospice or skilled-nursing facility care. 
Figure 10: Medicare spending declined with age for decedents over age 65 in 2014, mainly due to lower inpatient hospital spending

End-of-life patients do tend to be high-cost patients, and in general terms, that pattern seems appropriate. But I've written before that a main goal for end-of-life care, shared both by many patients and health care professionals, is to make greater use of hospice, skilled-nursing, and at-home care at the end of life, rather than intensive care units in a hospital setting. The evidence shows that over time, the costs of end-of-life care are a diminishing share of US health care spending, and it is consistent with the belief that a shift toward greater use of hospice and other options at the end of life is gradually underway.

Wednesday, July 20, 2016

Public Higher Ed: State Support Down, Tuition Up

State and local financial support for higher education is falling, and the share of costs covered by student tuition is rising. Perhaps not coincidentally the number of students enrolled in US public higher education is has fallen in the last few years. That's the evidence from the State Higher Education Executive Officers Association in it annual report, State Higher Education Finance 2015, released in April

The report notes: "In 2015, states invested $81.8 billion in higher education ... Local governments invested $9.1 billion from property tax revenue in 2015 primarily for local district community colleges." Here are some estimates over the last quarter-century, from 1990 to 2015, about the contributions of state and local spending on a per-student basis. The dollar figures are adjusted for inflation, so back in the early 1990s state and local spending on higher education was about $8,500 per student, but from 2011-2015 (despite a bump up in the last few years) it has been under $7,000 per student. Meanwhile, average tuition paid per student has more than doubled, from less than $3,000 back in the early 1990s to over $6,000 in 2015.

Putting those two trends together, it's no surprise that the share of public higher spending covered by tuition is rising. Indeed, this figure shows that the share has nearly doubled, from 25% back in 1990 to approaching 50% at present. The report discerns a pattern here: "Net tuition revenue per student tends to increase most rapidly during periods of recession, shifting more of the cost of higher education to students and families. ... During economic recessions, student share increases quickly and a new level is established during periods of recovery. Traditionally, the student share has not
declined significantly as state and local funding has been restored. It is likely that student share will surpass 50 percent during the next economic downturn."

Of course, rising student loans have helped students to pay the higher tuition during the past 25 years. But student loans now total $1.3 trillion, and the dip in higher education enrollments in the last few years (shown by the red line in the first figure) suggests that ever-higher tuition and loans are not the way to expand the effectiveness and enrollments in higher education.