Monday, March 8, 2021

Intergenerational Mobility and Neighborhood Effects

Raj Chetty delivered a keynote address titled "Improving Equality of Opportunity: New Insights from Big Data," to the annual meetings of the Western Economic Association International. It' has now been published in the January 2021 issue of Contemporary Economic Policy (39:1, 7–41, needs a subscription to access). The lecture gives a nice overview of some  of Chetty's work in the last few years. 

Chetty offers this nugget from previous research as as starting point: 
In particular, let’s think about the chance that a child born to parents in the bottom quintile—the bottom fifth of the income distribution—has of rising up to the top fifth of the income distribution. If you look at the data across countries, there are a number of recent studies that have computed statistics like this, called intergenerational transition matrices. You see that in the United States, kids born to families in the bottom fifth have about a seven and a half percent chance of reaching the top fifth. That compares with nine percent in the United Kingdom, 11.7% in Denmark, and 13.5% in Canada. ... One way you might think about this is that your chance of achieving the American dream are in some sense almost two times higher if you’re growing up in Canada rather than the United States. I think these international comparisons are motivating. It’s a little bit difficult to figure out how to make sense of them.
Start with a basic question: What do we need to  know if we want to measure intergenerational economic mobility? Basically, you need a measure of income for two separate generations, choosing an age when both older and younger generations are reasonably well-established in a career. Then you can compare where in the income distribution the younger generation falls, relative to their parents, and in that way have a sense of the chances of  how likely someone from one part of the income distribution is to move (higher or lower) to a different part of the income distribution. Best of all, you would like to have a really big nationally representative sample of this data, so that you can look at how such effects might vary across different locations, different sexes, and races/ethnicities. 

There is no single data set that will give you this kind of information. So Chetty and a team of researchers that has now grown to a couple of dozen people figured out how to combine existing datasets to get the necessary data.  Specifically, as Chetty describes it: 
In particular we use the 2000 and 2010 decennial census, as well as the American Community Survey—the ACS—which covers about 1% of Americans each year. We take that census data and link it to federal income tax returns from 1989 to 2015, creating a longitudinal data set. ... 

Those of you who have kids and live in the United States know that you have to write down your kid’s Social Security number on your tax return to claim him or her as a dependent. That allows us to link 99% of children in America back to their parents. What I’m going to describe here, the set of kids we’re seeking to study, are children born between 1978 and 1983 who were born in the United States or came to the United States as authorized immigrants while they were kids. That’s our target sample.

In practice, we have an analysis sample of about 20.5 million children once we’ve done the linkage of those various data sets, and that covers about 96% of the kidswe expect to be in our target sample. It’s not 100% because there are some kids for whom you’re not able to find a tax return or who weren’t claimed or who you weren’t able to link the census data to the tax data. But it’s still pretty good—at 96% you have essentially everybody you’re looking to study.

Just to be clear, this data is "de-identified": that is, stringent precautions are taken so that researchers don't see anyone's actual name or Social Security number or other personal information. Constructing this dataset is a substantial task, and it's a task that would not have been possible for researchers of previous generation. 

Here's one result. The horizontal axis shows the percentage of the income distribution for the older generation; the vertical axis the average for where the next generation ended up in the income distribution. The marked point shows that when the older generation was in the 25th percentile, the average outcome for the younger generation is the 41st percentile of the income distribution. Overall, the flatter this line, the more intergenerational income mobility exists: a perfectly flat line would mean that no matter where the older generation started, the expected result for the younger generation is the same. 

But while a lot of the previous research on intergenerational mobility had data that was nationally representative, the US-based research in particular has not had enough data that it could make plausible statements about intergenerational mobility at more local levels.

As a concrete example to help the exposition, Chetty sticks to this case where a parent's household is in the 25th percentile of the income distribution. Then he can ask: in different metropolitan areas across the United States, where is the average income of the younger generation higher or lower? 

Here's a map showing the pattern across the US where the younger generation is white and black men. The blue areas in the "heat map" show where the income of the younger generation is higher than the older generation; the white areas show where it's the same; the red areas show where it is lower. For black men, on the left, the overall pattern is mostly red and white: that is, the younger generation of black men usually have the same or lower income than their parents. For white men, on the right, the overall pattern is mostly blue and white; that is, the younger generation of white men are usually doing  the same or better than their parents.

There is of course lots to chew on in why these patterns differ across metropolitan area. For example, if one looks at the same graph for black women and white women, there is very little difference in this measure of intergenerational mortality. But the Chetty research group has so much data that they can look at much smaller geographic areas, including Census "tracts." There are about 70,000 tracts in the US that include about 4,000 people each. In certain high-population cities, the Chetty data lets a research look at intergenerational mobility at the level of a city block. 

Thus, Chetty and his team can tackle the question: what do patterns of intergenerational mobility look like not at the national level, or at the metro area  level, but at the neighborhood level. Sticking to the earlier concrete example of children born to a family at the 25th income percentile, does their income mobility differ depending on the neighborhood in which they live? Chetty writes: 
[T]he geographic scale on which we should think about neighborhoods as they matter for economic opportunity and upward mobility is incredibly narrow, like a half mile radius around your house. We find this not just for poverty rates, but many other characteristics. If you look at differences in characteristics outside that half mile radius, they have essentially no predictive power at all. I think that’s extremely useful from a policy perspective. We started this talk with the American dream. We now see that its origins, its roots, seem to actually be extremely hyperlocal.
Do the hyperlocal neighborhoods that have more intergenerational income mobility tend to share certain characteristics? Yes. For example, the share of households in the neighborhood with two-parent families makes a difference, as does the "social capital" of the neighborhood as measured by whether it has community gathering places like churches or even bowling alleys. 

At some level, all of this may sound like "common sense for credit," as economics has been described. Certain neighborhoods, not too far apart, are more correlated with intergenerational mobility than others.  But are the neighborhood effects something that could potentially be used for public policy? To put it another way, if the neighborhood changes, does someone from the younger generation basically have the same level of intergenerational mobility--because mobility is mostly determined by your family--or does someone from the younger generation perhaps have a higher or lower level of mobility--because the neighborhood has a separate causal effect on eventual economic outcomes?  

There are a variety of ways one might address this question. There's a major federal study called Moving to Opportunity. It was a randomized study: that is, a randomly selected half the sample got a certain program, but half didn't, so you can then compare outcomes between the results. Chetty describes: 
Let me start with the moving to opportunity (MTO) experiment. This experiment was conducted in five large cities around the United States. I’m going to focus on the case of Chicago, just to pick one illustration. In this experiment, researchers gave families living in very high poverty public housing projects, for instance, the Robert Taylor Homes in Chicago, one of two different kinds of housing vouchers through a random lottery. One group received Section 8 vouchers, which were vouchers that enabled them to move anywhere they could find housing. This assistance was on the order of about $1,000 per month in today’s dollars. The other group, called the experimental group, was given the same vouchers worth exactly the same amount, but they came with the restriction that users had to move to a low poverty area. These areas were defined as census tracts with a poverty rate below 10%.
It turns out that the randomly selected group who moved to lower-poverty areas as children did indeed have higher incomes as adults. The exact numbers of course have some statistical uncertainty built in. But as a bottom line, Chetty writes: 
That is to say, if I moved to a place where I see kids growing up to earn $1,000 more, on average, than in my hometown, I myself pick up about $700 of that. Just to put it differently, something like 70% of the variation in the maps that I’ve been showing you seems to reflect, if you take this oint estimate directly, causal effects of place, and 30% reflects selection. So a good chunk of it seems to actually be the causal effects of place.
Another statistical approach is just to look at families who move for any reason. Look at families with several children. When that family moves, some of their children will grow up in the new neighborhood for longer than others, and it turns out that the income gains from moving to the new neighborhood for children growing up in the same family line up with how long the child lived in that neighborhood. Chetty describes various other approaches to demonstrating that the neighborhood in which you grow up, where that is defined as the area within about a half-mile of your home, has a lasting effect on economic prospects.  

What public policy recommendations might flow from this finding? It would be impractical and costly to pay for vast numbers of people to relocate from current neighborhoods.  But there's a smaller-scale policy that might well be workable, which just involves providing information and connections to families with an interest in moving. Chetty writes:  
A different view is maybe this is about some sort of barriers, frictions that are preventing families from getting to these places. Maybe they lack information, maybe landlords in those neighborhoods don’t want to rent to them, maybe they don’t have the liquidity they need to get to those places, and so on.  We are conducting a randomized trial where we’re trying to address a bunch of those barriers by providing information, and by simplifying the process for landlords by providing essentially brokerage services, like search assistance, in the housing search process. We take that for granted in the high-end of the housing market, but it basically doesn’t exist at the low-end of the housing market. We take about 1,000 families, 500 of which receive the services, and 500 don’t, randomly chosen. ...
We found that this was an incredibly impactful intervention; we were extremely surprised by how much impact our services had on families’ likelihoods of moving to higher-opportunity neighborhoods. In the control group, less than one-fifth of families moved to higher opportunity areas. Eighty percent of families that received these vouchers chose to live in places that are relatively low mobility. In the treatment group, this completely changed. The vast majority of families in  the treatment group are now living in these high mobility places.I was just in Seattle talking to some of the families who’ve moved. They’re incredibly happy and describe how this small set of services, which only comes at a 2% incremental cost relative to the baseline cost of the housing voucher program, dramatically changed their choices and their kids’ experience. ...
There are a couple elements. First, from an economic perspective, we provide damage mitigation funds. This is basically an insurance fund that says that if anything goes wrong, we will cover it. In practice, the amount of expenses incurred are essentially zero, but I think it gives landlords some peace of mind. Second, there’s a simplification in the inspection process, which traditionally involves a lot of red tape and delays. We shortened the inspection process to 1 day, making it much simpler. Third—this actually surprised me—apparently telling landlords that their units can provide a pathway to opportunity for low-income kids actually makes them much more motivated to rent their units to certain families. ...
In fact, now we find landlords coming to the housing authority saying things like: “I heard about this program,” or, “I had a really good experience with your previous tenant, I want to now rent again.” I think we can change that equilibrium if we do it thoughtfully.

What about taking steps to improve the prospects for intergenerational mobility in neighborhoods in a direct way? It's worth remembering that Chetty's evidence suggests that what really matters is the half-mile around where people live, so what would seem to be called for is projects that improve the neighborhoods at a local level.  Chetty writes: 

What specific investments can be useful? Of course, that’s the question you’d want to know
the answer to. That could range from things like, most obviously, improving the quality of schools in an area to things like mentoring programs, and changing the amount of social capital, if we can figure out ways to measure and manipulate things like connectedness, reducing crime, and physical infrastructure. There are many such efforts that have been implemented over the years by local governments, nonprofits, and other practitioners.

You might ask which of those things is actually most effective; what’s the recipe for increasing upward mobility in a given place? I think the honest answer is that we just do not know yet. The reason for that is that there are lots of these place-based efforts where someone invests a lot of money in a given neighborhood. The neighborhood looks completely different 10 years down the road, but you have no idea whether that’s because new people moved in and displaced the people who were living there before, so the neighborhood gentrified, or if the people who lived there to begin with benefitted. And again, I think resolving that question comes back to having historical longitudinal data and being able to follow the people who lived there to begin with.
As you read this brief overview of a much larger body of work, you may find your mind raising questions about other possible connections or policies. That's natural. It's a genuinely exciting area of research that is opening up before our eyes. 

Friday, March 5, 2021

Debt and Deficits: Nostalgia for the 1980s

Back in the mid-1980s, the federal government under the Reagan administration ran what were widely considered to be excessive and risky budget deficits: from 1983 to 1986, the annual deficit was between 4.7% of GDP and 5.9% of GDP per year. The accumulated federal debt held by the public as a share of GDP rose from 21.2% of GDP in 1981 to 35.2% of GDP by 1987. I cannot exaggerate how much ink was spilled over this problem, some of it by me, back in those innocent and carefree time, before we learned to stop worrying and love the deficit.

The Congressional Budget Office has just released "The 2021 Long-Term Budget Outlook" (March 2021). There's nothing deeply new in it, but it made me think about how attitudes about budget deficits and government debt have evolved. The report notes: 
At an estimated 10.3 percent of gross domestic product (GDP), the deficit in 2021would be the second largest since 1945, exceeded only by the 14.9 percent shortfall recorded last year. ... By the end of 2021, federal debt held by the public is projected to equal 102 percent of GDP. Debt would reach 107 percent of GDP (surpassing its historical high) in 2031 and would almost double to 202 percent of GDP by 2051. Debt that is high and rising as a percentage of GDP boosts federal and private borrowing costs, slows the growth of economic output, and increases interest payments abroad. A growing debt burden could increase the risk of a fiscal crisis and higher inflation as well as undermine confidence in the U.S. dollar, making it more costly to finance public and private activity in international markets.
Here are a few illustrative figures. The first one shows accumulated federal debt over time since 1900. You see the bumps for debt accumulated during World Wars I and II, and during the Great Depression of the 1930s. If you look at the 1980s, you can see the Reagan-era rise in debt/GDP.  But after the debt/GDP ratio had sagged back to 26.5% by 2001, you can see the the big jump for debt incurred during the Great Recession, and then debt incurred during the pandemic recession, and then where the projections under current law would take us.
From an historical point of view, you can think of fiscal policy during the Great Recession and the pandemic recession as similar to what happened during the Great Recession and World War II. In both cases, there were two huge stresses within a period of about 15 years, and the federal government addressed both of them with borrowed money. In historical perspective, those Reagan-era deficits that caused such a fuss were just a minor speed bump. However, what's projected for the future has no US historical equivalent. 

This figure shows projections for annual budget deficits, rather than for accumulated debt. The figure separates out the amount of deficits that are attributable to interest payments in past borrowing (blue area). The "primary deficit" (purple area) is the deficit due to all non-interest spending. You'll notice that the primary deficit doesn't get crazy-high: it steadily grows from about 2.5% of GDP in the mid-2000s to 4.6% of GDP by the late 2040s. The problem is that the federal government gets on what I've called the "interest payments treadmill," where high interest payments are helping to create large annual deficits, and then large annual deficits keep leading to higher future interest payments. 
If the government could take actions to hold down the rise in the primary deficit over time, with some mixture of spending cuts and tax increases (or does it sound better to say spending "restraint" and tax "enhancements"?), then this could also keep the US government from stepping on the interest payments treadmill.  

This figure shows projected trends for spending and taxes, under current law. You can see the sepnding jump during the Great Recession, and then the jump during the pandemic recession. Assuming current law, projected tax revenues as a share of GDP don't change much going forward. However projected outlays do rise.

CBO explains the rise in outlays: 
Larger deficits in the last few years of the decade result from increases in spending that outpace increases in revenues. In particular:
  • Mandatory spending increases as a percentage of GDP. Those increases stem both from the aging of the population, which causes the number of participants in Social Security and Medicare to grow faster than the overall population, and from growth in federal health care costs per beneficiary that exceeds the growth in GDP per capita.
  • Net spending for interest as a percentage of GDP is projected to increase over the remainder of the decade as interest rates rise and federal debt remains high. 
There's been some talk in recent years about how, in a time of low interest rates, it could be an excellent time for the US government to make long-run investments that would pay off in future productivity. This case has some merit to me, but it's not what is actually happening. Instead, the fundamental purpose of the US government has been shifting. Back in 1970, about one-third of all federal spending was direct payments to individuals: now, direct payments to individuals are 70% of all federal spending. The federal government use to have missions like fighting wars and putting a person on the moon: now, it cuts checks. The CBO has this to say about the agenda of using federal debt to finance investments: 
Moreover, the effects on economic outcomes would depend on the types of policies that generate the higher deficits and debt. For example, increased high-quality and effective federal investment would boost private-sector productivity and output (though it would only partially mitigate the adverse consequences of greater borrowing). However, in CBO’s projections, the increasing deficits and debt result primarily from increases in noninvestment spending. Notably, net outlays for interest are a significant component of the increase in spending over the next 30 years. In addition, federal spending for Social Security, Medicare, and Medicaid for people age 65 or older would account for about half
of all federal noninterest spending by 2051, rising from about one-third in 2021.
For decades now, we have known that a combination of the aging of the post-World War II "baby boom" generation combined with rising life expectancies was going to raise the share of elderly Americans. We have also known for decades that primary programs for meeting the needs of this group--like Social Security and Medicare--have made promises that their current funding sources can't support. We have been watching US health care costs rise as a share of GDP For decades.  Meanwhile, the US economy has been experiencing slow productivity growth, which makes addressing all problems closer to a zero-sum game.  

Neither during the Great Recession of 2007-2009 nor during the heart of the pandemic recession in 2020 and early 2021 was an appropriate time to focus on the long-term future of government debt. But averting our eyes from the trajectory of the national debt is not a long-term strategy. 

Wednesday, March 3, 2021

Teaching Current Monetary Policy

Most of the ingredients of the standard intro econ class have been pretty stable for a long time: opportunity cost and budget constraints, supply and demand, perfect and imperfect competition, externalities and public goods, unemployment and inflation, growth and business cycles, benefits and costs of international trade. In contrast, the topic of how the central bank conducts monetary policy has shifted substantially more than decade ago, in a way that makes earlier textbooks literally obsolete. Jane Ihrig and Scott Wolla offer an overview of the changes in "Monetary Policy Had Changed, Has Your Instruction?" (Teaching Resources for Economics at Community Colleges Newsletter, March 2021).  For a more in-depth discussion, Ihrig and Wolla wrote: “Let’s close the gap: Revising teaching materialsto reflect how the Federal Reserve implements monetary policy” (October 2020, Federal Reserve Finance and Economics Discussion Series 2020-092)

(The TRECC Newsletter has financial support from the NSF and other sources. It currently has about 1500 subscribers. If you are are involved in community college teaching, or you teach a substantial number of lower-level undergraduate courses, you may want to put yourself on the subscription list.)

As Ihrig and Wolla point out, the traditional pedagogy of how monetary policy works focuses on how the Federal Reserve can raise or lower a specific targeted interest rate: the "federal funds" rate. It goes like this: 

One big change is in the "policy implementation" arrow. When I was taking intro econ in the late 1970s or when I was teaching big classes in the late 1980s and into the 1990s, the textbooks all discussed three tools for conducting monetary policy: open market operations, changing the reserve requirement, or changing the discount rate. 

Somewhat disconcertingly, when my son took AP economics in high school last year, he was still learning this lesson--even though it does not describe what the Fed has actually been doing for more than a decade since the Great Recession. Perhaps even more disconcertingly, when Ihrig and Wolla looked the latest revision of some prominent intro econ textbooks with publication dates 2021, like the widely used texts by Mankiw and by McConnell, Brue and Flynn, and found that they are still emphasizing open market operations as the main tool of Fed monetary policy. 

The Fed still targets the federal funds interest rate. But as for the earlier methods of policy implementation, there literally is no reserve requirement any more--no requirement that the banks  hold a certain portion of their assets on reserve at the Fed. The discount rate, where the Fed makes direct loans to institutions, has not been a commonly-used tool of monetary policy, because it was seen as a sign of financial weakness, perhaps even looming insolvency, if a bank needed could not get credit from other sources and needed to borrow from the Fed. The Fed still does open market operations for various reasons, but it is not the tool used to target the federal funds interest rate. 

What changed? In the Great Recession, the Fed had reduced its target interest rate--the federal funds interest rate--essentially to zero percent, as show in the figure. The Fed had hit the "zero lower bound," as economists say, but it still wanted and needed to try to boost the economy. 

In  particular, during the Great Recession there was deep uncertainty about the true value of certain mortgage-backed securities, and with very few buyers willing to purchase such securities, their value had plunged. Banks and financial institutions that were holding such securities as part of their assets were threatened with insolvency as a result.  Thus, the Fed started a program of "quantitative easing," where the Fed purchased both these mortgage-securities and also Treasury debt directly. Ihrig and Wolla describe it this way: 
This shift in framework was spurred on by actions during the global financial crisis. Between 2008 and 2014 the Fed conducted a series of large-scale asset purchase programs to lower longer-term interest rates, ease broader financial market conditions, and thus support economic activity and job creation. These purchases not only increased the Fed's level of securities holdings but also increased the total level of reserves in the banking system from around $15 billion in 2007 to about $2.7 trillion in late 2014. At this point, reserves were no longer limited but instead became quite plentiful, or "ample."
Bank reserves expanded enormously, as shown in this figure. In turn, this shift from "limited reserves" to the new era of "ample reserves" opened up the possibility for a new tool of monetary policy: the Fed could pay interest on these bank reserves. 
In short, the primary tool that the Fed now uses for the conduct of monetary policy is "interest on reserves," often abbreviated as IOR. This isn't hard to explain. Reserves go up and down for two reasons: one is if the Fed is using quantitative easing to buy Treasury debt or mortgage-backed securities directly, or not; the other is if banks prefer to hold reserves at the Fed and receive the IOR payments, or not. By moving the IOR up or down, the Fed still seeks to target the federal funds interest rate. For example, that rise in the fed funds rate from about 2016-2018, shown above, was a result of a rise in the interest rate paid on reserves.  

Thus, if you (or your textbook) is not emphasizing interest paid on reserves as the primary tool of monetary policy, or if (heaven forbid) you are still emphasizing open market operations as the main tool of monetary policy, you are more than a decade out of date. That said, here are some of my remaining questions about how to teach at the intro level how monetary policy is conducted. 

1) Should the previous pedagogy from before the ample-reserves era be eliminated from intro-level textbooks? The case for keeping it is partly that many intro-level books teach some recent economic history of the last few decades, and one can explain the shift to the "ample reserves" period. Also, some central banks around the world, like the People's Bank of China, still use tools like altering reserve reserve requirements. But as time passes, the case for dropping the earlier pedagogy grows. 

2) Although interest on reserves is the primary tool that the Fed has been using for targeting the Federal funds interest rate, there is also a secondary tool: the overnight reverse repurchase agreement facility that the Fed can use to put a floor under the federal funds interest rate, if it does not rise as desired with changes in the interest paid on reserves. Frankly, I despair of teaching the reverse repo market in an intro-level class. Given that the Fed does not see it as the primary tool, it's one of those topics where I would wait for an appropriate student question and be ready to talk about it during office hours. Or maybe it deserves a "box" in a textbook as some additional terminology worth mentioning. But maybe there's a smart teacher out there with a way to make this work for the intro-level class? 

3) Although interest on reserves it the Fed's main policy tool for normal times, neither of the last two recessions have been "normal." One big change mentioned above was the shift to quantitative easing. Another change is the shift to "forward guidance," or trying to affect current economic conditions by announcing the future path of Fed interest rate policy. Yet another is emergency lending facilities aimed at certain parts of financial markets, to make sure they will keep functioning even under economic stress. Here's Ihrig and Wolla in the Fed working paper: 
For example, during both crises, the Fed conducted large-scale asset purchases to either deliberately push down longer-term interest rates (the motive during the 2007-2009 financial crisis and subsequent recession) or aid market functioning and help foster accommodative financial conditions (the motive during the COVID-19 pandemic). As a traditional open market operation, when the Fed also made adjustments to existing lending facilities and introduced new, emergency lending facilities to help provide short-term liquidity to banks and other financial institutions. For example, the Fed expanded its currency swap program where it loans dollars to foreign central banks to alleviate dollar funding stresses abroad. It also introduced the Primary Dealer Credit Facility during both stress events, which provided overnight loans to primary dealers and helped foster improved conditions in financial markets.

The current standard practice for teaching the conduct of monetary policy at the intro level is still emerging. At present, my own take is to discuss the shift from limited reserves too ample reserves to give some sense of how these issues have evolved, but as time passes, it will eventually be time to drop the old limited-reserves approach entirely. Instead, for the current time, the tools of monetary policy to teach at the intro level would be interest on reserves, quantitative easing, forward guidance, and emergency (but temporary) lending facilities.

Tuesday, March 2, 2021

Putting Monetary Values on Health Costs of Coronavirus

W. Kip Viscusi delivered the Presidential Address at the (virtual) Southern Economic Association meetings last November on the subject "Economic Lessons for Coronavirus Risk Policies." The paper is forthcoming in the Southern Economic Journal; for now, it's available at the SSRN website a Vanderbilt University Law School Working Paper (Number 21-04, January 21, 2021). 

Viscusi is known for, as he says early in the paper, attempting to "strike a meaningful balance between risk and costs," even though "[e]conomic analyses in these domains are sometimes challenging and necessarily involve treading on controversial terrain. 

Thus, his analysis starts with standard estimates for the "value of a statistical life" of $11 million. The idea of a VSL is discussed here and here.  But the key point to understand is that the number is built on actual  real-world choices about risk, like how much more do people get paid for a riskier job, or what safety requirements for a car or a consumer product are judged to be "acceptable" or "too expensive." For example, say the government proposes a regulation that costs $110 million, but it is projected to reduce risks in a way that saves 11 lives in a city of 1.1 million people. That decision is implicitly saying that it's worthwhile for government to spend of $10 million per life saved. This reduction in risk is referred to as a "statistical" life saved. Using the standard measure of $11 million, mortality costs alone for COVID-19 in the US were $3.9 trillion through December 2020.

It's a little harder to use these same statistical methods to measure morbidity, rather than mortality, because morbidity covers everything from feeling crappy for a few days to a long and life-threatening hospital stay. But Viscusi uses some studies based, for example,  on what people are willing to spend to avoid hospitalization, and argues that the health costs of morbidity may add 40-50% to costs of mortality.

As one more health-related cost, Viscusi has long argued that at some high level of lost income (or high costs), the reactions to those lower income levels will also raise the risks of death, and he cites some of his own recent work to this effect: 
[A] loss in income of just over $100 million will lead to one expected death ...  This relationship highlights the important link between the performance of the economy and individual health. Shutting down economic activity to limit the pandemic has adverse health consequences in addition to the favorable reduction in risk due to increased social contacts. Income transfers such as unemployment compensation consequently serve a health-related function in dampening these losses.
Again, the value of a statistical life studies depend are not based on a social or political judgement about fairness or equity: they are based on empirical studies of what people are actually willing to pay to reduce certain risks (or what compensation they need to receive for taking on those risks). Thus, the calculations lead to Viscusi "treading on controversial terrain," as he puts it. 

For example, it is generally true that those with lower-incomes will need less compensation for taking on risk. He writes: 
There are strong income-related dependencies of the VSL. Many estimates of the income elasticity of the U.S. VSL are in the range of 0.6, and some available estimates are in excess of 1.0. Proper application of these income-related variations in the VSL would reduce the VSL applied to the poor. Many of those most affected by COVID-19 have below average income levels, including essential workers in grocery stores and other retail establishments, as well as those engaged in the production and delivery of food. Similarly, people whose jobs do not permit them to work at home and who must rely on public transit for their commute also are subject to considerable COVID-19 risks. Policies that protect their lives would be less highly valued if there were income adjustments. Given the strong
societal interest in maintaining the efforts of such individuals, it is unlikely that there would be support for applying a lower value to their lives, which in turn would lead to less policy emphasis on protecting their well-being.
Similarly, the value of a statistical life studies tend to find an inverse U-shape with age. Here's an illustrative figure from Viscusi, based on the empirical evidence: 

As Viscusi notes, the interpretation whether the empirical findings of a different VSL by age has led to political controversies.
Embarking on any age variation in the VSL is a controversial undertaking. In its analysis
of the Clear Skies Initiative, the U.S. Environmental Protection Agency (EPA) adopted a
downward age adjustment for those age 65 and older, reducing their applicable VSL by 37%. The result was a political firestorm against the “senior discount,” leading to headlines in critical articles such as “Seniors on Sale, 37% off” and “What’s a Granny Worth?” EPA abandoned this approach given the public outcry.
Notwithstanding the potential controversy that arises from even considering the prospect
of downward adjustments in the VSL for people who are older, what would the effect of
adopting an age-adjusted VSL be on the estimates of the mortality costs of COVID-19 Instead of a VSL of $11 million for the age 85, their VSL declines to $3 million. The average age-adjusted VSL for the COVID-19 age distribution is $6.3 million if there is no discounting of the VSLY stream and $5.2 million if a discount rate of 3% is applied to the stream of VSLY values. The total effect of accounting for age adjustments to the VSL is to cut the mortality cost estimate roughly in half.
To be clear, Viscusi is not advocating that public policy should place a different value on people's lives by age, income, or any other metric. He writes: 
[T]he risk equity concept that I have found to be attractive in many contexts is what I term `equitable risk tradeoffs.' Thus, rather than equalizing an objective measure of risk or life expectancy levels, the task is to adjust policies to set the risk-money tradeoff equal to a common population-wide VSL.
Some people find the VSL approach to thinking about risk just a useful way of codifying the choices that we are already making; others find it one more example of economics run amok. If you fall into the latter category, you might want to at least consider that when it comes to COVID-19 and future pandemics, Viscusi-style monetary values on mortality and morbidity offer a justification for a much more aggressive public spending response than the United States has actually done. As one example, Viscusi writes: 
Given the tremendous benefits that could be derived by having more adequate medical resources, it is preferable from a benefit-cost standpoint to make provision before health crises arise so that severe rationing is not required for the next pandemic. In anticipation of future pandemics, it is feasible to acquire high-quality ventilators at a cost from $25,000 to $50,000. Adding in the cost of medical support personnel would raise the annual  cost to about $100,000. A reserve supply of ventilators could be a component of an anticipatory pandemic policy. Preparing for future pandemics remains a cost-effective strategy even for annual probabilities of a pandemic on the order of 1/100. However, survey evidence by Pike et al. (2020) suggests that support for protective efforts of this type is unlikely to emerge, as there is a lack of public concern with long-term pandemic risks. As a result, there is likely to be a continued shortfall in preparations for prospective risks, leading to future repetitions of the difficult rationing decisions posed by COVID-19.
In addition, Viscusi points out that a number of medical ethicists in the last year have been talking about how to ration available health care resources in various ways, : that is, about who will get certain kinds of care and who won't. Viscusi-style estimates of the value of a statistical life make the case for high levels of government spending to avoid such rationing. As he writes: 
If human life is accorded an appropriate monetized value, the application of VSL and efficient principles for controlling risks will lead to greater levels of protection than will result if medical personnel follow the guidance provided by many prominent medical ethicists.

Monday, March 1, 2021

The Coming Evolution of Electric Power in the US

Even readers who are only experiencing the Texas electricity disruptions from afar may wish to consider a new report from an expert panel at the National Academy of Sciences, The Future of Electric Power in the United States (2021, prepublication copy downloadable for free).  Here's a summary of a few of the main economic  and technological changes facing the US electricity industry. 

1) A Potentially Large Increase in Demand for Electricity 

In the last couple of decades, per capita demand for electricity has been fairly flat in the US, while per capita demand for energy has actually declined. This is in part because of greater energy efficiency, and also in part because the US economy has been shifting to service-oriented production that is uses less energy for each dollar of output produced. 

But looking ahead, at present about 0.1% of energy for US transportation come from electricity. If the market for electric cars expands dramatically, this will clearly rise--and given that much of the recharging of electric vehicles would happen in homes, the transmission of electricity to residences could rise substantially. Also at present, "12 percent of energy used by industry came from electricity in 2019, almost all deep-decarbonization studies suggest that electrification of certain industrial end uses will be important."  More broadly, pretty much all aspects of the digital economy run on electricity. One estimate is that for the world as a whole,  "data centers and networks consumed around 380 terawatt hours (TWh) of electricity in 2014–2015, about 2 percent of total electricity demand."  

Of course, continual increases in efficiency of electricity use--say, by large household appliances and new light bulbs--could offset this rise in demand for electricity. On the other side, changes in more people using electricity in their homes or all purposes, rather than using natural gas or oil, could also shift electricity demand higher. 

2) A Shift in How Electricity is Generated 

Even if the demand for electricity doesn't risk, there is an ongoing push toward generating electricity in different ways that produce less or no carbon emissions. As the report says: "An economy that decarbonizes is an economy that electrifies." Here's a graph showing the current sources of electricity in the US. 
Obviously, coal and natural gas still dominate electricity generation in the US. As the report notes: "So far, the United States is making modest progress in decarbonizing its power sector primarily
through the replacement of coal with natural gas—sometimes called `shallow decarbonization'—but the path to deep decarbonization is not yet widely agreed upon." 

There doesn't seem to be much of a push for more nuclear or hydro (although there probably should be), which means that we're talking about here is expanding the green and yellow areas to account for a much larger share of electricity generation over the next 2-3 decades. Maybe other technologies like carbon capture and storage or hydrogen can make a difference, too. My sense is that most people (and politicians) have no realistic sense of what would be needed to make solar and wind the main elements of the electrical grid. It's not just expanding wind and solar by a factor of maybe 10, with all that implies for building additional solar and wind facilities, but it's also building the new transmission lines needed to get the power where we want it to be and building the electricity grid so that it can deal with the intermittent nature of these sources of electricity. 

3) The Changing Grid Edge

The old traditional method of producing and selling electricity involved a few big generating facilities:  the report calls this the "make it-move it-use it" model. The new method involves decentralized electricity generation from a variety of possible sources, right down to rooftop solar panels that can supply electricity to the grid. In addition, electricity providers are gaining the capability to charge different amounts at different times of day, or for different levels of household usage, or even to cycle your air conditioning or heat on and off at times when demand is especially high. There may also be rising demand for different kinds of electricity provision: for example, homes may want "fast-charging" capability for their electric car, but status quo electricity elsewhere in the grid. Certain areas may have industrial or information-technology facilities that need especially large quantities of electricity. Issues of physical security of the grid and also cybersecurity of the grid are important, too.  

Factors like these mean that the electricity grid itself is becoming enormously more complex to manage, with a much wider range of options for generation, distribution, and usage.  

The report discusses an array of other topics, including jobs that will be lost and gained, global supply chains for the kinds of equipment that will be needed (much of it based in China), effects for people with lower-incomes who may have less access to reliable electricity, and so on.  

But in my mind, perhaps the key point is recognizing that the evolution of the electricity industry is heavily shaped by physical realities and by regulations at both state and federal levels. With many industries, I'm happy to let private companies struggle in the market to get my business, with a relatively mild background of government regulation. But for most of us, electricity is delivered from a common  grid, and I don't have a choice of competing providers. The rules for  how my power company will buy power, and how it determines what it will charge me, are heavily regulated and largely opaque to the public. Questions about where new generating facilities and new transmission lines will be built often seem to involve one separate dispute at a time, rather than as part of an overall strategy. The incentives for a given electricity utility to invest in research and development or to experiment with different kinds of service provision,, given the realities of regulation, can be quite limited. 

Ins short, the electricity market is facing needs for transformational change in a regulatory and economic context where change has traditionally been constricted and piecemeal. It's all too easy to imagine how pressures for change and limits on change will collide. 

Thursday, February 25, 2021

India: Pivoting from the Pandemic to Economic Reforms

Each year, the Economic Division in India's Ministry of Finance publishes the Economic Survey of India (January 2021). The first volume is a set of chapters on different topics: the second volume is a point-by-point overview of the last year's developments, in fiscal, monetary, and trade policy, along with developments in main sectors like agriculture, industry, and services. Here, I'll cherry-pick some points that caught my eye in looking over the first volume. 

Of course, any discussion of a country's economy 2020 will start with the pandemic. All statements about what "worked" or "didn't work" during 2020 are of course subject to revision as events evolve. As a country with many low-income people living in high-density cities, and high absolute numbers of elderly people, India was clearly a country with what looked as if it might experience large health costs in the pandemic. But the report argues that at least for 2020, India's COVID-19 response worked well. (For those not used to reading reports from India, "lakh" refers to 100,000, and a "crore" is 100 lakh, or 10 million.)

India was amongst the first of the countries that imposed a national lockdown when there were only 500 confirmed cases. The stringent lockdown in India from 25th March to 31st May was necessitated by the need to break the chain of the spread of the pandemic. This was based on the humane principle that while GDP growth will come back, human lives once lost cannot be brought back. 

The 40-day lockdown period was used to scale up the necessary medical and para-medical infrastructure for active surveillance, expanded testing, contact tracing, isolation and management of cases, and educating citizens about social distancing and masks, etc. The lockdown provided the necessary time to put in place the fundamentals of the '5 T' strategy - Test, Track, Trace, Treat, Technology. As the first step towards timely identification, prompt isolation & effective
treatment, higher testing was recognized as the effective strategy to limit the spread of infection. At the onset of the pandemic in January, 2020, India did less than 100 COVID-19 tests per day at only one lab. However, within a year, 10 lakh tests were being conducted per day at 2305 laboratories. The country reached a cumulative testing of more than 17 crore in January, 2021. ... 

The districts across India, based on number of cases and other parameters were classifie into red, yellow and green zones. Across the country, ‘hotspots’ and ‘containment zones’ were identified – places with higher confirmed cases increasing the prospect of contagion. This strategy was increasingly adopted for intensive interventions at the local level as the national lockdown was eased. ... 

India was successful in flattening the pandemic curve, pushing the peak to September. India managed to save millions of ‘lives’ and outperform pessimistic expectations in terms of cases and deaths. It is the only country other than Argentina that has not experienced a second wave. It has among the lowest fatality rates despite having the second largest number of confirmed cases. The recovery rate has been almost 96 per cent. India, therefore, seems to have managed the health aspect of COVID-19 well.

India's economy seems to have experienced a V-shaped recession, with a sharp decline during the 40-day lockdown period but then a return to pre-pandemic levels by the end of 2020. 

Other chapters in the report look at other issues that have become more salient as a result of the pandemic. For example, India's economy has labored for years under what has been called the "license raj," referring back to the British colonial period for a metaphor to describe how an extraordinarily instrusive level of licensing and regulation limits flexibility and growth in the India's economy. 

Element of the "license raj" still exist. As one example, the report notes: 

International comparisons show that the problems of India’s administrative processe derive less from lack of compliance to processes or regulatory standards, but from overregulation. ... [T]he issue of over-regulation is illustrated through a study of time and procedures taken for a company to undergo voluntary liquidation in India. Even when there is no dispute/ litigation and all paperwork is complete, it takes 1570 days to be stuck off from the records. This is an order of magnitude longer than what it takes in other countries. ... 

The ‘World Rule of Law Index’ published by the World Justice Project provides cross-country comparison on various aspects of regulatory enforcement. The index has various sub-categories, which capture compliance to due processes, effectiveness, timelines, etc. In 2020, India’s rank is 45 out of 128 countries in the category of ‘Due process is respected in administrative proceedings’ (proxy for following due process). In contrast, in the category ‘Government regulations are effectively enforced’ (proxy for regulatory quality/effectiveness), the country’s rank is 104 (Table 1). India stands at 89th rank in ‘Administrative Proceedings are conducted without unreasonable delay’ (proxy for timeliness) and 107th in ‘Administrative Proceedings are applied and enforced without improper influence’ (proxy for rent seeking).

Another example looks back at some aftereffects of policies taken during the Great Recession back in 2008-2009. During that time, India's banking and financial regulators instituted a policy of "forbearance," meaning that they wouldn't crack down on financial institutions that were in a shaky position during a deep recession. This policy can make sense in the short-term: if regulators crack down on banks during a recession, it can propagate a deeper recession. But soon after the recession, this policy of forbearance needs to stop--and in India that's not what happened.  

During the GFC [global financial crisis], forbearance helped borrowers tide over temporary hardship caused due to the crisis and helped prevent a large contagion. However, the forbearance continued for seven years though it should have been discontinued in 2011, when GDP, exports, IIP [international investment position] and credit growth had all recovered significantly. Yet, the forbearance continued long after the economic recovery, resulting in unintended and detrimental consequences for banks, firms, and the economy. Given relaxed provisioning requirements, banks exploited the forbearance window to restructure loans even for unviable entities, thereby window-dressing their books. The inflated profits were then used by banks to pay increased dividends to shareholders, including the government in the case of public sector banks. As a result, banks became severely undercapitalized. Undercapitalization distorted banks’ incentives and fostered risky lending practices, including lending to zombies. As a result of the distorted incentives, banks misallocated credit, thereby damaging the quality of investment in the economy. Firms benefitting from the banks’ largesse also invested in unviable projects. In a regime of injudicious credit supply and lax monitoring, a borrowing firm’s management’s ability to obtain credit strengthened its influence within the firm, leading to deterioration in firm governance. The quality of firms’ boards declined. Subsequently, misappropriation of resources increased, and the firm performance deteriorated. By the time forbearance ended in 2015, restructuring had increased seven times while NPAs [non-performing assets] almost doubled when compared to the pre-forbearance levels.

But with these kinds of ongoing issues duly noted, India has also seized the opportunity of the pandemic to carry out some long-promised structural reforms. For example, one change is that farmers are now allowed to sell their crops to anyone, anywhere, rather than being required to sell only to a designated local agency.  Another issue of long-standing is that India has long offered a range of subsidies to smaller firms,  which sounds OK until you realize that if a small firm thinks about growing into a larger firm, it realizes that it would lose its government subsidies. These kinds of labor regulations have been substantially loosened, and the number of regulations pared back. 

The increase in the size thresholds from 10 to 20 employees to be called a factory, 20 to 50 for contract worker laws to apply, and  100 to 300 for standing orders enable economies of scale and unleash growth. The drastic reductions in compliance stem from (i) 41 central labour laws being reduced to four, (ii) the number of sections falling by 60 per cent from about 1200 to 480, (iii) the maze due to the number of minimum wages being reducing from about 2000 to 40, (iv) one registration instead of six, (v) one license instead of four, and (vi) de-criminalisation of several offences.
In the next few years, it will be interesting to see if these changes make a real difference, or if they have just rearranged the furniture, with the same regulatory burden reconfigured. 

Another aftereffect of the pandemic is to raise the visibility of public health programs in India. These were already on the rise. For example,  there "an increase in public [health care] spend from 1 per cent to 2.5-3 per cent of GDP – as envisaged in the National Health Policy 2017 – can decrease the Out-Of-Pocket Expenditures from 65 per cent to 30 per cent of overall healthcare spend." There are programs to expand telemedicine and the infrastructure needed to support it. 

Also, India's government launched a program in 2018 aimed at providing more access to health care (which is mostly privately provided in India) to the low-income population. 
In 2018, Government of India approved the Ayushman Bharat Pradhan Mantri Jan Arogya Yojana (AB-PM-JAY) as a historic step to provide healthcare access to the most vulnerable sections in the country. Beneficiaries included approximately 50 crore individuals across 10.74 crores poor and vulnerable families, which form the bottom 40 per cent of the Indian population. The households were included based on the deprivation and occupational criteria from the Socio-Economic Caste Census 2011 (SECC 2011) for rural and urban areas respectively. The scheme provides for healthcare of up to INR 5 lakh per family per year on a family floater basis, which means that it can be used by one or all members of the family. The scheme provides for secondary and tertiary hospitalization through a network of public and empanelled private healthcare providers. It also provides for three days of pre-hospitalization and 15 days of posthospitalization expenses, places no cap on age and gender, or size of a family and is portable across the country. It covers 1573 procedures including 23 specialties (see Box 1 for details). AB-PM-JAY also aims to set up 150,000 health and wellness centres to provide comprehensive primary health care service to the entire population.
Finally, in India as in so many countries, there is often a policy question as to whether the country should be striving for additional economic growth or for a reduction in inequality: or more specifically, what the tradeoffs would be in prioritizing one of these goals over the other. The Survey looks at potential tradeoffs and data across the states. It finds that in the context of India, there doesn't seem to be a conflict 

[T]he Survey examines if inequality and growth conflict or converge in the Indian context. By examining the correlation of inequality and per-capita income with a range of socio-economic indicators, including health, education, life expectancy, infant mortality, birth and death rates, fertility rates, crime, drug usage and mental health, the Survey highlights that both economic growth – as reflected in the income per capita at the state level –and inequality have similar relationships with socio-economic indicators. Thus, unlike in advanced economies, in India economic growth and inequality converge in terms of their effects on socio-economic indicators. Furthermore, this chapter finds that economic growth has a far greater impact on poverty alleviation than inequality. Therefore, given India’s stage of development, India must continue to focus on economic growth to lift the poor out of poverty by expanding the overall pie. Note that this policy focus does not imply that redistributive objectives are unimportant, but that redistribution is only feasible in a developing economy if the size of the economic pie grows.

For some previous posts on India's economy, see:

The first link discussed a three-paper "Symposium on India" in the Winter 2020 issue of the Journal of Economic Perspectives (where I work as Managing Editor). 

Wednesday, February 24, 2021

Robert J. Gordon: Thoughts on Long-Run US Productivity Growth

Leo Feler has a half-hour interview with Robert J. Gordon on "The Rise and Fall and Rise Again of American Growth"  (UCLA Anderson Forecast Direct, February 2021, audio and transcript available). The back-story here is that Gordon has been making the argument for some years now that modern economic interventions, like the rise of information technologies and the internet, have not had and will not have nearly the same size effect on productivity as some of the major technologies of the past like the spread of electricity or motor vehicles (for some background, see here and here). 

Here, Gordon makes a distinction worth considering between growth in productivity and growth in consumer welfare.
Let’s divide the computer age into two parts. One is the part that developed during the 1970s and 80s and came to fruition in the 1990s, with the personal computer, with faster mainframe computers, with the invention of the internet, and the transition of every office and every business from typewriters and paper to flat screens and the internet, with everything stored in computer memory rather than filing cabinets. That first part of the computer revolution brought with it the revival of productivity growth from the slow pace of the 70s and 80s to a relatively rapid 2.5% to 3% per year during 1995 to 2005. But unlike the earlier industrial revolution where 3% productivity growth lasted for 50 years, this time it only lasted for ten years. Most businesses now are doing their day-to-day operations with flat screens and information stored in the cloud, not all that different from how they did things in 2005. In the last 15 years, we’ve had the invention of smartphones and social networks, and what they’ve done is bring enormous amounts of consumer surplus to everyday people of the world. This is not really counted in productivity, it hasn’t changed the way businesses conduct their day-to-day affairs all that much, but what they have done is change the lives of citizens in a way that is not counted in GDP or productivity. It’s possible the amount of consumer welfare we’re getting relative to GDP may be growing at an unprecedented rate.
To understand the distinction here, say that you pay a certain amount for access to television and the internet. Now say that over time, the amount of content you can access in this way--including shows, games , shopping , communication with friends, education, health care advice, and so on--rises dramatically, while you continue to pay the same price for access. In a productivity sense, nothing has changed: you pay the same for access to television and internet as you did before. But from a consumer welfare perspective, the much greater array of more attractive and easier-to-navigate choices means that you are better off. 

The expression "timepass" is sometimes used here. One of the big gains of information technology is that, for many people, it seems like a better way of passing the time than the alternatives. 

Gordon also points out that the shift to working from home and via the internet could turn out to involve large productivity gains. But as he also points out, shifts in productivity--literally, producing the same or more output with fewer inputs--is an inherently disruptive process for the inputs that get reduced. 
This shift to remote working has got to improve productivity because we’re getting the same amount of output without commuting, without office buildings, and without all the goods and services associated with that. We can produce output at home and transmit it to the rest of the economy electronically, whether it’s an insurance claim or medical consultation. We’re producing what people really care about with a lot less input of things like office buildings and transportation. In a profound sense, the movement to working from home is going to make everyone who is capable of working from home more productive. Of course, this leaves out a lot of the rest of the economy. It’s going to create severe costs of adjustments in areas like commercial real estate and transportation.
When asked about how to improve long-run productivity, Gordon's first suggestion is very early interventions for at-risk children:  
I would start at the very beginning, with preschool education. We have an enormous vocabulary gap at age 5, between children whose parents both went to college and live in the home and children who grow up in poverty often with a single parent. I’m all for a massive program of preschool education. If money is scarce, rather than bring education to 3 and 4 year olds to everyone in the middle class, I would spend that money getting it down as low as age 6 months for the poverty population. That would make a tremendous difference. ... This isn’t immediate. These children need to grow into adults. But if we look out at what our society will be like 20 years from now, this would be the place I would start.
For some of my own thoughts on very early interventions, well before a conventional pre-K program, see here, here and here