(1) DURATION DEPENDENCE AND LABOR MARKET CONDITIONS

    Kory Kroft, Fabian Lange, and Matthew J. Notowidigdo

The likelihood of receiving a callback for a job interview sharply declines with unemployment duration.
----------------------------------------------------------------------

According to a recent report by the Congressional Budget Office, long-term
unemployment may "produce a self-perpetuating cycle wherein protracted spells of
unemployment heighten employers' reluctance to hire those individuals, which in
turn leads to even longer spells of joblessness." Policymakers and researchers
alike tend to believe that this adverse effect of a long spell of unemployment
undermines the smooth functioning of the labor market and entails large social
costs. Economists refer to the phenomenon as "negative duration dependence."
In Duration Dependence and Labor Market Conditions: Theory and Evidence from a
Field Experiment (NBER Working Paper No. 18387), authors Kory Kroft, Fabian
Lange, and Matthew Notowidigdo confirm that the likelihood of receiving a
callback for a job interview sharply declines with unemployment duration. This
effect is especially pronounced during the first eight months after becoming
unemployed. Their estimates suggest that this effect is quantitatively
important, and that duration dependence is stronger when jobs are relatively
abundant. These results imply that employers statistically discriminate against
workers with longer unemployment durations and that employer screening plays an
important role in generating duration dependence.
To study duration dependence, the authors submitted fictitious resumes to real,
online job postings in each of the 100 largest metropolitan areas in the United
States, and then tracked "callbacks" from employers for each submission. In
total, they "applied" to roughly 3,000 job postings in Sales, Customer Service,
Administrative Support, and Clerical job categories, submitting roughly 12,000
resumes. The resumes they created characterized the "applicant's" employment
status and, if unemployed, the length of the current unemployment spell, which
ranged from 1 to 36 months and was randomly assigned. As a result, this
experiment directly uncovered duration dependence arising through employers'
beliefs about unemployed workers.
--Lester Picker

http://papers.nber.org/papers/W18387?utm_campaign=dig&utm_medium=email&utm_source=dig

(2) CHANGES IN SMOKING AND OBESITY AFFECT FUTURE LIFE EXPECTANCY

    Samuel H. Preston, Andrew Stokes, Neil K. Mehta, and Bochen Cao

Life expectancy for 40-year old men in 2040 is expected to increase by 0.92 years compared to... 2010. For women [there is] ... a smaller increase in expected life expectancy -- only 0.26 years.
----------------------------------------------------------------------

Deaths from smoking and obesity play a significant role in any estimates of
future U.S. life expectancy. Cigarette consumption per adult per year has fallen
from a high of more than 4,000 in the early 1960s to fewer than 2,000 in the
early 2000s, and that reduction in smoking should increase life expectancy.
However, population obesity began to rise in the 1980s, and that trend is
associated with decreased life expectancy. To further complicate any estimates
of the effects of these trends on mortality, researchers observe that men and
women behave differently and thus will be affected differently by changes in
smoking and obesity.
In Projecting the Effect of Change in Smoking and Obesity on Future Life
Expectancy in the United States (NBER Working Paper No. 18407), co-authors
Samuel Preston, Andrew Stokes, Neil Mehta, and Bochen Cao forecast the likely
net effect of changes on U.S. mortality rates from 2010 through 2040. They find
that men benefit from significantly reduced smoking, with life expectancy for
40-year old men in 2040 expected to increase by 0.92 years compared to life
expectancy in 2010. For women, the more muted decline in smoking is largely
offset by the surge in obesity, resulting in a smaller increase in expected life
expectancy -- only 0.26 years.
The authors note that data from National Health and Nutrition Examination
Surveys suggest that the probability of becoming obese, given one's weight at
age 25, has been nearly constant for the past 18 years. Given that figure, they
estimate that 47 percent of men and 51 percent of women may be obese by 2040.
Furthermore, by 2020 and thereafter a majority of obese women will be morbidly
obese -- that is, with a body mass index above 35. These changes are expected to
raise death rates for 40-to-84 year-old men by as much as 13 percent and for
women of the same ages by as much as 20 percent.
Using deaths from smoking-related cancers as a proxy for smoking, the authors
estimate that age-specific death rates for male smokers will decline between
2010 and 2040, largely because the heaviest smoking male cohorts were more than
80 years old in 2010. They predict that men will gain 1.5 years of life by 2040
from reduced smoking. Since women began smoking more recently than men, the
heaviest smokers among women do not "age out" until 2025-30. Therefore, life
expectancy gains for women as a whole rise after 2025, cumulating at 0.85 years
for 40-year-old women in 2040.
--Linda Gorman

http://papers.nber.org/papers/W18407?utm_campaign=dig&utm_medium=email&utm_source=dig

(3) EXPERIMENTAL EVIDENCE FROM ENERGY CONSERVATION PROGRAMS

    Hunt Allcott and Todd Rogers

At the end of four years, the average household that was still receiving [energy use] reports was taking actions equivalent to turning off a standard 60-watt light bulb for about 11 hours each day.
----------------------------------------------------------------------

In 2008, one electric utility implemented a program in which it randomly
selected about 43,000 of its household customers to receive "Home Energy
Reports" which feature personalized feedback, social comparisons, and energy
conservation information. About a quarter of that sample group was randomly
selected to no longer receive those reports after two years. In The Short-Run
and Long-Run Effects of Behavioral Interventions: Experimental Evidence from
Energy Conservation (NBER Working Paper No. 18492), Hunt Allcott and Todd Rogers
analyze the short-term and long-term results of this program on energy
conservation behavior.
They find that in the short run, there is a pattern of "action and backsliding."
As the initial reports arrive, some consumers are immediately motivated to
conserve. They adjust thermostats, turn off lights, and unplug unused
electronics. However, those households soon begin to "backslide" toward their
pre-intervention levels of energy use, at least until the arrival of subsequent
reports, which induce at least some of these same households to conserve.
The authors find that the effects of this program are highly durable. As long as
the reports continue to arrive over the third and fourth years, the households
receiving them continue to incrementally reduce their energy use. Even for
households dropped from the program after year two, about two-thirds of the
effect on conservation remains two years later. At the end of four years, the
average household that was still receiving reports was taking actions equivalent
to turning off a standard 60-watt light bulb for about 11 hours each day; the
average household dropped from the program was still taking actions equivalent
to turning off that light bulb for 7.5 hours each day.
However, the evidence from this experimental program suggests that the
households that continued to receive reports were no more inclined than
households in the control group (no reports) to invest in physical capital, such
as energy efficient appliances or insulation. Based on self-reported data on
actions taken to conserve energy, it would seem that the program does not
significantly induce households to adopt new energy habits, but instead might
motivate households to do more of the same things that they were already doing
to reduce energy consumption. The authors conclude that it is important to
repeat an intervention until participants have developed a new "capital stock"
of habits or other technologies.
--Claire Brunel

http://papers.nber.org/papers/W18492?utm_campaign=dig&utm_medium=email&utm_source=dig

(4) DYNAMIC ASPECTS OF FAMILY TRANSFERS

    Kathleen McGarry

Parents appear more motivated to help their less fortunate children than to treat all of their children equally.
----------------------------------------------------------------------

In Dynamic Aspects of Family Transfers (NBER Working Paper No. 18446), Kathleen
McGarry examines data on 17 years of transfers between parents and their adult
children and finds that parents evaluate their children's income prospects
before making gifts. When those prospects dim, especially after a reversal such
as the loss of a job or a divorce, parents are willing to kick in money to ease
the situation.
"Transfers made in conjunction with specific events in the child's life appear
to be important and suggest that parents frequently respond to negative shocks
to the child's income," she writes. McGarry also finds that gifts to adult
children both vary over time and differ across children in the same family. In
directing their gifts, parents appear more motivated to help their less
fortunate children than to treat all of their children equally.
Previous studies have found that in any given year, parents give their adult
children an estimated $65 billion (in 2010 dollars). Some of these studies
suggested that the flows could be explained by an altruism model, which predicts
that parents give more to their neediest children. Other studies seemed to
suggest that the exchange model, which says that parents pay for services
rendered by their children, is what is at work. McGarry develops a dynamic model
of parental altruism, built around the idea that a parent's expectations of a
child's income change over time and that giving changes with it. She points out,
however, that some giving appears motivated by factors outside this model.
Drawing on data from the Health and Retirement Study from 1992 to 2008, she
provides some of the first evidence on how parental giving varies over time. For
example, approximately 14 percent of the children in her sample of 3,776
families received a cash transfer from their parents in a particular two-year
period, but only 6 percent received a transfer in two adjacent surveys.
Similarly, nearly half of the children (46 percent) received a transfer in one
of the nine biennial surveys, but less than 1 percent reported receiving a
transfer in all of those periods. Moreover, the period-to-period changes in
transfers were strongly correlated with changes in the child's income.
The gifts to children within a family didn't average out over time. Those with
the smallest current income, and those with low permanent income, tended to get
the largest amounts from parents.
McGarry also finds evidence that parents give for positive developments in a
child's life. Those who completed their sixteenth year of schooling between two
surveys (and thus who likely graduated from college) received approximately $500
more than those who did not. Those who married between one survey and the next
had a roughly 30 percent greater probability of getting a transfer than their
counterparts who did not. The birth of a grandchild offered the biggest bonus of
all. The parent of a grandchild reported an average transfer of $5,758, more
than $1,500 greater than the average of $4,236 for those who didn't report a new
child.
Some of the largest transfers came after negative events that affected a child's
income. The loss of a job resulted in an average transfer of $5,257, the largest
of any life event except the birth of a grandchild. Those who went through a
divorce had a 61 percent greater probability of receiving money from parents
than those who didn't get divorced.
--Laurent Belsie

http://papers.nber.org/papers/W18446?utm_campaign=dig&utm_medium=email&utm_source=dig

(5) EMISSIONS, ELECTRIC CARS, AND OTHER POLLUTION-CONTROL POLICIES

    Joshua S. Graff Zivin, Matthew Kotchen, and Erin T. Mansur

Moving to increased use of PEVs may not reduce CO2 because of local variation in daily, and even hourly, patterns of electricity production and consumption.
----------------------------------------------------------------------

The policy goal of reducing CO2 and other emissions from the transportation
sector has led to interest in plug-in electric vehicles (PEVs). The United
Kingdom has made electric cars an important part of its overall efforts to
reduce carbon-dioxide emissions. In California, policymakers have required
manufacturers to offer PEVs for sale as part of its own overall climate-change
measures. Elsewhere in the United States, the federal government and other state
governments offer an array of financial incentives to promote electric cars,
including tax credits for consumers.
PEVs shift the energy consumption associated with transportation toward
centrally-generated electrical power and away from engines in individual
vehicles, leading to more efficient energy consumption. But power plants also
have emissions. In fact, electric power plants, particularly those that burn
coal, are an important source of carbon-dioxide emissions, accounting for
example for more than 40 percent of domestic CO2 emissions in the United States.

How greater use of PEVs would affect CO2 emissions in the United States depends
critically on the way that the electricity used to charge the batteries in these
cars is generated. In Spatial and Temporal Heterogeneity of Marginal Emissions:
Implications for Electric Cars and Other Electricity-Shifting Policies (NBER
Working Paper No. 18462), Joshua Graff Zivin, Matthew Kotchen, and Erin Mansur
conclude that "electricity shifting" policies, such as moving to increased use
of PEVs, may not reduce CO2 because of local variation in daily, and even
hourly, patterns of electricity production and consumption. The analysis is
complicated by the fact that electric power is often transmitted across long
distances through power lines, so that the CO2 emissions associated with
charging a PEV in one state may be located in another.
Using data from the U.S. Environmental Protection Agency, the Energy Information
Administration, the Federal Energy Regulatory Commission, and other sources, the
authors analyze energy use and emissions over a three-year period, from 2007
through 2009. They track energy production and consumption by the hour of the
day within different regions. They conclude that on net, in the western United
States and in Texas, driving PEVs would result in lower carbon-dioxide emissions
than driving fuel-efficient hybrid cars. But in other regions, such as the upper
Midwest, where the fuel mix for electricity generation is more heavily tilted
toward coal, the charging of PEV batteries during the recommended hours of
midnight to 4 AM could result in more emissions than those associated with the
average car now on the road. "Underlying this result is a fundamental tension
between load management of electricity and achieving environmental goals," the
authors conclude. "The hours when electricity is the least expensive to produce
tend to be the hours with the greatest emissions."
--Jay Fitzgerald

http://papers.nber.org/papers/W18462?utm_campaign=dig&utm_medium=email&utm_source=dig

(6) AFRICAN AGRICULTURAL DECISIONS AFTER RELAXING CREDIT AND RISK CONSTRAINTS

    Dean Karlan, Robert Darko Osei, Isaac Osei-Akoto, and Christopher Udry

When provided with insurance against the primary catastrophic risk they face -- drought or floods -- farmers are able to find the resources to increase expenditure on their farms.
----------------------------------------------------------------------

In recent decades lagging agricultural productivity in sub-Saharan Africa has
motivated charities and policymakers to implement assistance programs, usually
by promoting hybrid seeds and fertilizers. But in focus group interviews,
farmers in the region commonly report "lack of money" or concerns regarding the
high risk from weather and disease as key obstacles to investing in their land.
In Agricultural Decisions after Relaxing Credit and Risk Constraints (NBER
Working Paper No. 18463), co-authors Dean Karlan, Robert Osei, Isaac Osei-Akoto,
and Christopher Udry analyze the results of experiments conducted in northern
Ghana in which farmers randomly received cash grants, opportunities to purchase
rainfall index insurance, or a combination of the two. They find that when
provided with insurance against the primary catastrophic risks they face, the
farmers are able to find the resources to increase expenditures on their farms.
The authors seek to understand how capital and risk interact, and under what
circumstances they lead to underinvestment. To do this, they devise a three-year
random experiment. In year one, maize farmers are provided a cash grant (or no
cash grant) and a rainfall insurance grant (or no rainfall insurance grant). In
year two, a cash grant is again offered, but rainfall insurance is no longer
given out for free. Instead, it is for sale at randomly varied prices ranging
from one eighth of its actuarially fair price to the market price (that is,
actuarially fair plus a market premium to cover servicing costs). In year three,
the farmers are offered only the insurance pricing experiment.
The researchers find that agricultural investment responds quite a bit to the
rainfall insurance grant, but relatively little to the cash grants. The salient
constraint to farmers' investment appears to be uninsured risk: when provided
with insurance against the primary catastrophic risk they face -- drought or
floods -- farmers are able to find the resources to increase expenditure on
their farms. Even at the actuarially fair price for insurance, 40 to 50 percent
of the farmers demand it, and they purchase coverage for more than 60 percent of
their cultivated acreage. Because the farmers do not seem to completely trust
that payouts will be made when rainfall insurance trigger events occur, the
demand for insurance is sensitive to the experience of the farmer himself and to
that of others in his social network with insurance. Demand for insurance
increases after either the farmer or others in his network receive an insurance
payout, and demand decreases if a farmer was previously insured and the rainfall
was good, resulting in no payout.
The authors suggest that their experiment provides an important lesson for the
microcredit community: capital constraints are not the only, or sometimes even
the most important, hurdle to raising investment. Risk is a key hindrance to
investment and thus to improved income and growth. Mitigating risk even without
an infusion of capital leads to higher investment.
--Matt Nesvisky