Full text of Econ Focus (Federal Reserve Bank of Richmond) : Third Quarter 2024, Volume 29, Number 3

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

FEDERAL RESERVE BANK OF RICHMOND

THIRD QUARTER 2024

Retirement Savings

Stormy weather ahead?

Fed Lending After the
2023 Bank Crisis

Inside
Food Banks

Interview with
Raghuram Rajan

VOLUME 29 ■ NUMBER 3
THIRD QUARTER 2024

Econ Focus is the economics
magazine of the Federal Reserve
Bank of Richmond. It covers
economic issues affecting the
Fifth Federal Reserve District
and the nation and is published
by the Bank’s Research Department.
The Fifth District consists of the
District of Columbia, Maryland,
North Carolina, South Carolina,
Virginia, and most of West Virginia.
DI R EC TO R O F P U B L ICATI ONS

Renee Haltom

FEATURE

8 RETHINKING RETIREMENT SAVINGS

Many Americans have surprisingly little set aside for retirement. Why, and what can
be done to boost their nest eggs?

14 FOOD BANKS: LIFELINES TO THOSE IN NEED

For millions, food banks fill a crucial gap. How do they do it and just how big is the
need they address?

EDI TO R

David A. Price
MA N AG IN G E D ITO R

Lisa Davis
STA F F WR ITE R S

Tim Sablik
Matthew Wells
EDI TO R IA L A SSO C IATE

Katrina Mullen

CON TR IB U TO R S
Charles Gerena
Brooke Hansbrough
Julian Kikuchi
Stephanie Norris
Sam Louis Taylor
Laura Dawson Ullrich
Sonya Ravindranath Waddell
Zhu Wang
DESI G N

Janin/Cliff Design, Inc.
PUB L IS H E D BY
the Federal Reserve Bank
of Richmond
P.O. Box 27622
Richmond, VA 23261
www.richmondfed.org
www.twitter.com/RichFedResearch

Subscriptions and additional copies:
Available free of charge through our website at
www.richmondfed.org/publications.

Reprints: Text may be reprinted
with the disclaimer in italics below. Permission
from the editor is required before reprinting
photos, charts, and tables. Credit Econ Focus and
send the editor a copy of the publication in which
the reprinted material appears.
The views expressed in Econ Focus are those of
the contributors and not necessarily those of the
Federal Reserve Bank of Richmond or the Federal
Reserve System.
ISSN 2327-0241 (Print)
ISSN 2327-025x (Online)

DEPARTMENTS
1 PRESIDENT’S MESSAGE

Flexible Work and Women’s Participation

2 UPFRONT

New from the Richmond Fed’s Regional Matters Blog

3 AT THE RICHMOND FED
Connecting Women Economists

4 FEDERAL RESERVE

Central Bank Lending Lessons from the 2023 Bank Crisis

12 RESEARCH SPOTLIGHT

Skill Mismatch, Layoffs, and Bouncing Back

14 POLICY UPDATE

Independence, If You Can Keep It

18 ECONOMIC HISTORY
Milton Friedman, Dissenter

22 INTERVIEW

Raghuram Rajan on leading a central bank and creating a digital payment system

27 DISTRICT DIGEST

Preparing to Work: Changing Demand for Postsecondary Education?

32 OPINION

Artificial Intelligence: Potentials and Prospects

Cover Image: Getty Images

Scan here to
subscribe to
Econ Focus

PRESIDENT’S MESSAGE

Flexible Work and Women’s Participation

W

hen COVID-19 hit in 2020,
one of the many shocks families faced was the closing of
schools and child care centers. In many
families, the burden of dealing with
such shocks was disproportionately
borne by the mom — so this sudden
change hit women’s labor force participation hard. Commentators labeled it a
“she-cession.”
But in the time since, women’s participation in the workforce has bounced
back. Among women aged 25 to 54,
the labor force participation rate — as
economists call it — actually reached
an all-time record in May and remains
high at 77.9 percent (versus 77 percent
in February 2020). Moreover, as
Brookings Institution researchers have
noted, women with young children, ages
4 and under, have led the charge, with
their labor force participation growing
at the fastest rate. With the upcoming
start of the new school year, it’s a good
time to ask: What happened?
Part of the story is the reopening of
schools and a rebound in the supply
of child care workers — which fell
dramatically during the pandemic but
has since climbed back to its former
level of staffing. From what I’ve been
seeing and hearing, though, that’s only
one piece of what’s been going on.
Of course, hybrid work and work
from home has been a significant
change from four years ago. For a large
part of the population, balancing work
and home is or seems more manageable
than before. In principle, that opens
up labor force participation to men and
women alike — but in practice, as I
noted, it’s most often women who are
juggling these multiple tasks.
Perhaps the growth of hybrid work
and work from home explains why the

recovery in labor force participation
among mothers of young children has
been most pronounced among women
with a college degree: As my Research
Department colleagues have observed,
those jobs are the ones most likely to
have flexible work options.
The other piece of the puzzle I find
interesting is the place where participation has strongly dropped and stayed
lower: people 65 and up. This runs
counter to the long-term trend. Before
the pandemic, employment among older
adults had been rising gradually for
several decades. The reversal may have
been driven in part by pandemic health
concerns and by pandemic-related
workplace changes that made retirement look more attractive. Another
important factor may have been the
run-up in asset prices, including housing prices, making retirement appear
financially safer than it had before.
One hypothesis that’s been shared
with me, which resonates with stories I
hear from my contemporaries, is there

are a lot of older people taking care
of their grandchildren, supplementing their children two or three days a
week. That, too, helps women return
to work or keep working while maintaining some balance. These grandparents have left the formal economy but
are supplementing the costly and often
hard-to-access child care industry by
working informally.
In a tight labor market, employers
invested in flexibility and the resulting
increase in participation has in turn
helped bring labor supply and demand
into better balance. But the demographic outlook for the workforce looks
to be a continuing challenge, as fertility has dropped, leaving the classes
currently in K-12 ever-smaller over
time. To move the needle further, I
increasingly hear employers considering investments in reducing barriers
to work, like housing subsidies, on-site
child care, or tailored roles attractive
to older workers.
We should also acknowledge that
no one canceled the business cycle, so
these employer initiatives haven’t yet
been tested by a downturn. What will
we see when that happens? Will we see
a pullback in flexibility as firms face
less pressure on hiring and retention?
If so, will women’s participation absorb
an outsized brunt of the contraction?
Or will business leaders keep their eye
on the long term and continue to invest
in initiatives to bring people into the
workforce?

Tom Barkin
President and Chief Executive Officer

econ focus

• third quarter • 2023 1

UPFRONT
b y k at r i n a m u l l e n

New from the Richmond Fed’s Regional Matters blog
Joseph Mengedoth. “Have Some Rural Areas
Turned the Tide on Population Decline?”
Between 2010 to 2020, more than half of Fifth District counties in
rural areas or small towns experienced population declines. Yet nearly
half of these counties went on to experience population growth from
2020 to 2023. The reversal in population growth is largely attributed
to domestic in-migration, which accounted for about 88 percent of
total net migration in these counties. The largest absolute gains were
in North Carolina: Rutherford County
attracted the greatest number of people
(2,392), followed by Nash County (1,968).
Possible reasons for the population growth
could be proximity to metro areas, flexible
work arrangements, and affordability — but
there are also the natural amenities in some
counties, which could attract more retirees.
Laura Dawson Ullrich and Stephanie
Norris. “Following the Money: State and
Local Funding for Community Colleges in
the Fifth District.”
Community colleges mostly rely on funding
from state and local appropriations as well
as tuition revenue and federal financial aid.
The annual state budget process determines
funding amounts, and each state uses a
different approach to distribute the funds.
Most Fifth District states divide their
higher education budget based on both institution type and full-time
equivalent, or FTE, enrollment, but it can still vary: South Carolina does
not use a formula, whereas Maryland uses a very specific formula that
gives community colleges nearly 30 percent of the state funding amount.
For local funding, community colleges in just 29 states receive money,
but this also varies from small amounts in Virginia to large amounts in
Maryland.
Sonya Ravindranath Waddell. “Employment Change:
Are Workers Coming or Going?”
In May, the Richmond Fed monthly business surveys asked firms about
how their employment, hiring, and separations have changed in the last
month. The share of firms reporting an increase in hiring (25 percent)
was higher than the share reporting an increase in voluntary separations
(14 percent) or involuntary separations (19 percent). Moreover, in a
tight labor market, many firms are not only backfilling open positions,
they are also turning to automation and outsourcing. Some 37.9 percent
of respondents reported implementing technology to automate tasks
2

econ focus

• third quarter • 2024

previously completed by employees, and 20.7 percent outsourced work
that was not previously outsourced.
Emily Wavering Corcoran and Sonya Ravindranath Waddell.
“Automation and AI: What Does Adoption Look Like for
Fifth District Businesses?”
Before automation or artificial intelligence (AI) can provide labor or total
factor productivity improvements, they must be adopted. The Richmond
Fed monthly business surveys recently asked
Fifth District firms about their adoption of
automation, including generative AI. Of the
respondents, 46 percent had automated
tasks in the past two years, and the majority
indicated that they plan to within the next
two years. While manufacturing firms were
more likely than service sector firms to have
implemented automation, the reason seems
generally to be to complement workers,
not to replace them. On the other hand,
while service sector firms were less likely
overall to adopt automation, they were
more likely to use AI in their automation
than manufacturing firms. Overall, the
responses indicate that it is the early days of
AI adoption, but the responses are helpful to
measure and understand this technology.
Surekha Carpenter and Adam Scavette.
“Understanding Immigration in the Fifth District: Where Did
International Migrants Settle?”
According to the Census Bureau, in 2022, international migration
to the United States returned to pre-COVID-19 levels, and recent
estimates indicated that immigration has surged to unprecedented
levels. The Fifth District’s population change and migration, however,
differed from those of the nation as a whole between 2020 and 2023.
Compared to the United States overall, the Carolinas experienced a
higher rate of population growth, while the District of Columbia and
West Virginia had negative growth. Virginia was close to the national
average, while Maryland remained flat. In the Carolinas, most of this
growth is attributed to domestic migration, as they have especially
attracted residents who have moved from across the United States.
Between 2020 to 2023, however, 90 percent of Fifth District counties
had net positive international migration, particularly in urban counties
and counties in and near metro areas. Some rural and small-town
areas in the Fifth District — especially homes to universities — also
experienced higher net international migration. EF

AT THE RICHMOND FED
by charles gerena

Connecting Women Economists

i m age : f e d era l r e s e rv e ba n k o f r i c h mo nd

W

ithin competitive fields like economics, informal
Such events are another avenue for connection, which is
connections are important for career advancewhy the center will host its own conference in Richmond
ment. To help create these valuable connections, the this November. “We are aiming to organize this conference
Richmond Fed’s Center for Advancing Women in Economics
in a way that is inclusive and appealing to all, especially
launched a fellowship program this year as part of its
women,” says Arantxa Jarque, a senior policy economist at
multifaceted approach to raising the visibility of women
the Richmond Fed and the center’s associate director. “We
economists.
also hope to bridge the gap between academia and policy,
“There is a leaky pipeline in the research track in
especially around the areas that the Fed is interested
economics, and this fellowship seeks to address it,”
in.” This fall, the center will also co-organize the annual
explains Marina Azzimonti, a senior economist and
Women in System Economic Research Conference with the
research advisor at the Richmond Fed
Atlanta Fed and Kansas City Fed.
who leads the center. Women received
Another way that the center will help
only one-third of new Ph.D.s in economics
women economists forge new connections
in 2022 and comprise just one-quarter of
is by compiling the Directory of Women
tenure-track economists at Ph.D.-granting
in Economics. “There are people within
institutions. The picture is similar at the
the profession who are aware of the lack
Fed’s regional Reserve Banks, where only
of diversity and try to proactively include
21 percent of senior-level economists are
underrepresented groups,” said Azzimonti
women. “Our hope is that women stay in
when recalling her career on the Speaking
the profession and are successful so they
of the Economy podcast. “But these
become leaders.”
groups may not be within their existThe annual fellowship is open to
ing networks, since they are less likely to
early-career researchers with a Ph.D. in
have met them in grad school or to have
economics or a related field. An intercrossed paths with them at conferences.
nal committee at the Richmond Fed
Oliko Vardishvili, one of the inaugural fellows It’s like a vicious cycle.”
selected the two inaugural fellows for
of the Center for Advancing Women in
The directory currently includes women
2024: Stephanie Johnson of Rice University Economics, participated in the Richmond Fed's economists with a Ph.D. who work in the
and Oliko Vardishvili of the University
CORE Week in May 2024.
Federal Reserve System. It is expanding to
of California, Irvine. Applications for the
include women economists at universities
2025 fellowship will be considered from junior researchwithin the Fifth District this year and eventually will cover
ers working on topics of interest to the Bank, including, but the entire United States.
not limited to, macroeconomics and central banking.
The directory will help put women economists on the radar
Johnson and Vardishvili will attend two of the Richmond in various ways, according to Jarque. For example, a reporter
Fed’s CORE Weeks, where they can connect with the
can search the directory for subject matter experts. An econoacademic researchers who are invited to collaborate with
mist can use it to recruit keynote speakers for a conference or
the Bank’s research economists throughout the year. So far, referees for a journal article. Or a Ph.D. student who is thinkthe networking opportunities offered during CORE Week
ing about working at the Fed can use it to contact women
conferences “have been invaluable for my professional
economists with similar research interests.
growth,” says Vardishvili, as has the hands-on assistance
Along with the Women in Macroeconomics Conference
she has received from the Bank’s economists.
at the University of Chicago and similar events for other
Fellows attend other events and present at one of the
areas of economics, plus mentoring and networking groups,
Richmond Fed’s online brown bag seminars, which provide
is this too much of a good thing for women economists? “In
additional opportunities to network as well as gather feedacademia, there is no such thing as too many opportunities
back on their ongoing research. “I have received very
to discuss your research or to learn from the people doing
tailored feedback about packaging my presentations and
frontier work,” notes Jarque.
tailoring my research papers to meet the specific requireThe Richmond Fed also stands to benefit, adds Jarque.
ments of target journals,” says Vardishvili. “The key is not
“Learning about the work of women ensures we can inform
only to have great research ideas and methodologies, but
policymaking with diverse perspectives that are more
also to learn how to effectively communicate your research.” likely to represent the communities we strive to serve.” EF
econ focus

• third quarter • 2024 3

FEDERAL RESERVE
by tim sablik

Central Bank Lending Lessons from
the 2023 Bank Crisis
The Fed moved quickly to support the financial system during a banking
panic last spring. Now, policymakers are evaluating what they learned.

I

n the spring of 2023, a pair of fast-moving bank runs threatened to spark a
widespread financial panic. On March
9, the 16th largest bank in the country, Silicon Valley Bank (SVB) in Santa
Clara, Calif., lost a quarter of its deposits
in a single day. It was set to lose another
62 percent of deposits the following day
before it was closed by regulators. On
March 10, New York-based Signature
Bank experienced a similarly rapid flight
of 20 percent of its deposits. It was closed
by regulators on March 12.
At the time of their collapses, SVB
($209 billion in assets) and Signature
Bank ($110 billion in assets) were the
second- and third-largest bank failures
in U.S. history. Their failures were also
exceptionally quick by modern standards. By comparison, Washington
Mutual, the largest bank failure in
American history, lost 10 percent of its
deposits over the course of 16 days in
September 2008.
The business models of SVB and
Signature Bank differed, but both
were hit by rapidly rising interest rates
following the post-pandemic surge in
inflation. Both also had a large share of
institutional depositors with accounts
that exceeded the Federal Deposit
Insurance Corporation (FDIC) insurance limit of $250,000, making the
depositors more likely to withdraw
funds at signs of trouble. The rapid failures of SVB and Signature Bank raised
concerns that other banks with similar
risks might soon follow.
THE CRISIS AND RESPONSE
In the days surrounding the failures
of SVB and Signature Bank, depositors
4

econ focus

• third quarter • 2024

fled banks with assets between $50
billion and $250 billion, moving their
money primarily to larger institutions.
According to a May 2024 paper by
Marco Cipriani and Thomas Eisenbach
of the New York Fed and Anna Kovner,
research director of the Richmond
Fed, a total of 22 banks experienced
runs last March.
The turmoil would ultimately claim
one more victim, First Republic Bank
in San Francisco, which began experiencing a run on March 10 and failed
on May 1. With $213 billion in assets,
it took the number two slot on the list
of largest bank failures, surpassing
SVB. According to a report from the
Group of Thirty, an independent global
body of economic leaders and experts
who advise on issues facing policymakers and market participants, the three
failed banks collectively held more
assets than all bank assets lost in the
2008 financial crisis.
As in that previous crisis, the Fed
acted swiftly to prevent financial turmoil
from sweeping up other institutions.
Borrowing at the Fed’s discount window,
a standing facility that makes short-term
loans to qualified banks, spiked from
$4.6 billion on March 9 to $152.9 billion
on March 15. The Fed also created an
additional lending facility on March
12 to support the financial system: the
Bank Term Funding Program (BTFP).
Through the BTFP, the Fed made loans
to banks in exchange for government
bonds and agency securities as collateral.
(See chart.)
These actions fulfilled one of the Fed’s
oldest functions: to serve as a “lender
of last resort” to the financial system.
Partly thanks to this intervention,

widespread failures were averted despite
many banks experiencing significant
stress. In the year since, Fed policymakers and academic researchers have been
examining the events of last March for
lessons on how to improve the central
bank’s lender-of-last-resort facilities
before the next crisis.
ROLE OF THE LENDER OF
LAST RESORT
By the nature of their business, banks
are susceptible to panics. They take
customer deposits, which can be
withdrawn on demand, and invest in
longer-duration assets like loans. Such
assets are often held to maturity and
may not be easy to sell quickly. If too
many depositors seek to withdraw
their money at once, a bank may not
have enough cash to meet the sudden
surge in demand. This can lead to a
run, as depositors rush to get their
money out while the bank still has
funds to pay them.
Financial regulators have sought to
ensure that banks are resilient against
runs by requiring them to hold enough
capital to absorb losses as well as
enough liquid assets to meet a sudden
surge in depositor demand. These
precautions must be balanced against
the fact that requiring banks to raise
more capital and hold more cash could
limit their capacity to make loans and
channel credit to productive uses in
the economy.
When a crisis eventually comes,
solvent but temporarily illiquid banks
can borrow from the Fed to weather
the storm. Even if central bank lending doesn’t ultimately prevent a bank’s

Borrowing from the Lender of Last Resort
Loans to banks from the Fed's discount window and Bank Term Funding Program
180
160
140
$BILLIONS

120
100
80
60
40
20
0

Jan.
2023

March
2023
Discount Window

May
2023

July
2023

Sept.
2023

Nov.
2023

Jan.
2024

March
2024

May
2024

BTFP

NOTE: Discount Window refers to loans from the Fed's Primary Credit facility.
SOURCE: Board of Governors of the Federal Reserve System

failure, it can avert the need for the
bank to sell assets at fire-sale prices
to meet depositor demand. Such sales
can fan the flames of the financial
panic by devaluing the assets held by
other institutions, potentially bringing
the run to their doors as well. Having
an entity to play this role in the U.S.
economy was a major motivation for
the creation of the Fed in 1913. In the
mid-19th and early 20th century, when
America had no central bank, banking
crises were frequent occurrences.
The discount window has been the
Fed’s primary lender-of-last-resort tool
since its founding. Banks pledge collateral — which can include loans, bonds,
and other asset-backed securities —
and the Fed determines the amount
of money the bank can borrow. (This
is typically the value of the collateral
minus a haircut.) While this facility was created to help the banking
system in an emergency, historically
banks have been reluctant to use it
even in a crisis.
That’s because borrowing from the
lender of last resort is often interpreted
as a sign that a bank has exhausted all
other options. Many bankers worry
that sending such a signal could
further intensify pressures for a run by

revealing the bank is in a weaker condition than its depositors may have realized. At a March event hosted by the
Brookings Institution’s Hutchins Center
on Fiscal and Monetary Policy, William
Demchak, the CEO of PNC Financial
Services Group, remarked, “The day
you hit it [the discount window] for
anything other than a test you effectively have told the world you failed.”
THE STIGMA CHALLENGE
Banks that borrow from the discount
window, then, would prefer to keep
that fact a secret. The Dodd-Frank Act
of 2010 requires the Fed to disclose the
identities of discount window borrowers after a two-year lag. In theory,
the revelation should come long after
the crisis has passed. But in practice,
market participants can often infer the
identities of discount window borrowers much sooner.
The Fed publishes weekly data
disclosing the assets and liabilities
of each Reserve Bank — including
discount window loans. Banks borrowing from the discount window do so at
their regional Reserve Bank. A spike
in lending at one of the 12 Federal
Reserve districts can therefore provide

a clue about which banks might have
borrowed based on where they are
headquartered. In 2020, the Fed made
some modifications to how it reports
this data to further mask individual
banks’ discount window activity. But in
an April article, Steven Kelly, the associate director of research at the Yale
Program on Financial Stability, argued
that it is often still possible to detect
a spike in certain borrowers’ discount
window use from the weekly reports.
“The Fed’s data does offer some
degree of obfuscation, but not enough,”
says Kelly. “The way that data is set up,
it’s the mid-sized and larger banks that
are most vulnerable to being revealed.
So when you have a crisis primarily
among mid-sized banks, like we did in
March 2023, there was a very real fear
of tapping the discount window and
being discovered by the market.”
In part because of this stigma,
banks have often turned to alternative
sources of emergency credit, including other Fed facilities. During the
financial crisis of 2007-2008, the Fed
created the Term Auction Facility
(TAF) as an alternative program
for making loans to banks. Unlike
discount window loans, the rates
on TAF loans were determined by
auction. This auction design may have
made it more difficult for the market
to deduce the identity of borrowers,
reducing the stigma banks faced when
borrowing from the Fed.
In 2023, borrowers from the Fed’s
BTFP may have also sought to avoid
discount window stigma. But in addition, says Huberto Ennis, group vice
president for macro and financial
economics at the Richmond Fed, “the
BTFP was designed to address a very
specific problem that some banks were
experiencing” — namely, the problems
associated with rapidly rising interest
rates.
The runs at SVB, Signature Bank,
and First Republic Bank were exacerbated by the fact that all three
held assets in the form of long-dated
securities that lost value when interest rates rose abruptly in 2022. The
econ focus

• third quarter • 2024 5

A LENDER OF NEXT-TO-LAST
RESORT
Created by Congress in 1932, the FHLB
system was set up to provide funding for mortgage lenders to support
the housing market during the Great
Depression. There are 11 FHLBs
that each serve a particular region.
Depository institutions can become
a member of their regional FHLB
and receive loans (called advances)
in exchange for eligible collateral.
While FHLB advances were originally
intended to support housing, banks
have used them as a source of general
liquidity in times of financial crisis.
This practice has led some to call the
FHLB system a “lender of next-to-last
resort.”
In the lead-up to the 2023 banking crisis, SVB, First Republic, and
Signature Bank all borrowed heavily from their FHLBs. According
to a March report from the U.S.
Government Accountability Office,
SVB and First Republic were the largest borrowers from the San Francisco
FHLB at the start of the year, and
Signature Bank was the fourth-largest
borrower from the New York FHLB.
6

econ focus

• third quarter • 2024

In March 2023, Silicon
Valley Bank lost a
quarter of its deposits in a single day; it
was quickly shuttered
by regulators and
later acquired by First
Citizens Bank. At the
time of its collapse,
it was the secondlargest bank failure in
U.S. history.

All three banks sharply increased their
borrowing and requests for FHLB
advances in early March as they experienced distress. For example, the
balance of FHLB advances to SVB
increased by 50 percent — from $20
billion to $30 billion — between March
1 and March 8.
While having an additional lender of
last resort during a crisis may seem like
a good thing, researchers have identified some issues with the FHLBs playing this role. In principle, FHLBs make
advances only to sound institutions in
exchange for good collateral. But in
practice, they may not always have the
strongest incentives to assess borrower
soundness because their collateral
requirements make it unlikely that they
would lose money if the institution
fails, according to Columbia University
law professor Kathryn Judge.
In a May 2014 article in the Cornell
Law Review, Judge wrote that “no
FHLBank has ever lost money on an
advance despite the failure of many
banks with significant outstanding
advances.” If financial firms can obtain
funding from the FHLBs that the
market would otherwise not provide
them, they can delay their reckoning until their ultimate failure is much
larger and costlier to the financial
system. This could contribute to excessive risk-taking by failing firms, which
have a greater incentive to take on
more risk to avoid failure.
Another problem identified by
researchers is that, unlike the Fed,
FHLBs need to raise funding from

the market to issue advances. Since
marketplace funding takes time to
execute, the ability of FHLBs to lend
could become constrained precisely
when they are needed to act as a
lender of last resort.
“The Federal Home Loan Banks
simply aren’t as capable emergency
lenders as the Fed, particularly when
it comes to large sums, because they
have to raise the money,” says Kelly.
“FHLBs can also be procyclical in a
way that the Fed is not. During crises,
FHLBs have raised the haircuts they
apply to collateral, or, as we saw in
the case of First Republic, they may
suddenly stop lending to a bank to
figure out what is going on. Those are
things that the Fed doesn’t do.”
A third challenge is that borrowing from FHLBs can complicate a
bank’s ability to also borrow from the
Fed. When a bank borrows from the
discount window, it needs to put up
collateral without competing claims,
allowing the Fed to seize it if the
bank fails to repay the loan. When
FHLBs issue advances, they impose a
lien on the collateral that supersedes
all other claims, making it ineligible
for use at the discount window. This
can be cleared up with discussions
between the Fed and FHLBs, but in
a fast-moving crisis there may not be
enough time. In the case of Signature
Bank, FDIC Chair Martin Gruenberg
said in congressional testimony that
these issues were only resolved with
“minutes to spare before the Federal
Reserve’s wire room closed.”

i m age : ge tt y / i sto c k , su n d ry ph oto g ra ph y

FE DE R AL R E S E RVE

BTFP accepted high-quality long-dated
assets (such as Treasuries and U.S.
agency mortgage-backed securities) as
collateral at their face, or par, value.
“This allowed banks to receive
cash from the Fed for the amount of
government-guaranteed securities,”
explains Ennis. “If banks used those
securities as collateral to borrow from
the discount window, they would have
been discounted based on their market
value.”
Even so, Cipriani, Eisenbach, and
Kovner found that banks were reluctant
to borrow from either Fed channel in
March 2023. All 22 banks that experienced runs relied on borrowing to meet
depositor demand, but only some chose
to borrow from the discount window or
the BTFP. In contrast, all 22 borrowed
from their Federal Home Loan Bank
(FHLB).

SPEED AND READINESS
The speed of the March 2023 crisis
also revealed important lessons for
policymakers. In the aftermath of the
financial crisis of 2007-2008, regulators
introduced a new requirement known
as a Liquidity Coverage Ratio (LCR),
which requires banks of a certain size
to hold highly liquid assets proportionate to their total assets. (See “Liquidity
Requirements and the Lender of Last
Resort,” Econ Focus, Fourth Quarter
2015.) The LCR presumes that during
a run, between 25 percent and 40
percent of a bank’s large uninsured
deposits could flee over the course of a
month.
“With SVB, we saw the attempted
withdrawal of over 60 percent of
deposits in one day,” says Darrell
Duffie, a professor of management and
finance at Stanford University. “It is
clear now, if it wasn’t before, that large
uninsured depositors will move their
funds out of a bank that’s in trouble
very quickly, particularly financially
savvy large depositors who are going
to be attuned to these risks.”
Short of having enough liquidity on
hand to meet such a rapid and large
deposit flight, the 2023 crisis suggests
the importance of banks being
prepared to borrow from the lender
of last resort at a moment’s notice.
All three banks that failed experienced difficulties borrowing from the
discount window, in part due to a
lack of practice with the requirements
involved. SVB had not tested its ability
to borrow from the discount window
at all in 2022, and Signature Bank had
not conducted such tests in the five
years before its failure.
In a 2021 Richmond Fed Economic

Brief, Ennis found that in the noncrisis
period of 2010-2017, very few institutions with less than $1 billion in assets
borrowed from the Fed’s discount
window: only 7 percent of domestic
banks and 2 percent of credit unions.
Starting this year, the Fed has begun
releasing annual statistics on banks’
and credit unions’ readiness to borrow
from the discount window. Between
2022 and 2023, the number of institutions signed up to use the discount
window increased by 9.4 percent, from
4,952 to 5,418. Ennis says that to the
extent that the events of March 2023
revealed that banks were not fully
informed about the steps they needed
to take to be ready to borrow quickly
from the discount window, it is helpful
for the Fed to share information and
create greater awareness.
“At the same time, I would say that
there should be no presumption that a
bank needs to be able to borrow from
the discount window,” he says. “Banks
need to make that determination
themselves after considering all the
relevant information.”
Last year’s crisis also cast a spotlight on the Fed’s readiness to handle
requests that could come at any time in
the fast-paced era of modern finance.
In a 2023 article, Yale Program of
Financial Stability Executive Fellow
Susan McLaughlin noted that there
are different cutoff times for pledging collateral at the discount window
to borrow that same day. These cutoff
times can be as early as 9:15 a.m.
Pacific Time depending on the type of
securities being pledged, and two of the
failed banks were located on the West
Coast. This is why the Fed recommends
that banks pre-position their collateral
at the discount window to be ready to

borrow right away in an emergency. In
the wake of last year’s crisis, some have
called for this pre-positioning to be
taken a step further.
POTENTIAL REFORMS
A January report from the Group of
Thirty’s Working Group on the 2023
Banking Crisis, chaired by former New
York Fed President William Dudley,
recommended that the Fed require
banks to pre-position enough collateral
at the discount window to cover all
their runnable liabilities, which would
notably include all uninsured deposits.
“It would mitigate the risk of runs
triggered merely because one depositor thinks other depositors are going to
move,” says Duffie, who was an adviser
on the report.
Fed officials have indicated they are
looking at such a change. In a May
speech, Michael Barr, the vice chair
for supervision on the Fed’s Board of
Governors, said the Fed was considering
requiring banks to pre-position collateral
at the discount window based on a fraction of their uninsured deposits.
Barr also acknowledged criticisms
about the technology and procedures
surrounding discount window borrowing and the need to reduce stigma.
“Given the important role of the
discount window, we’re also actively
working to improve its functionality,”
he said. In March, the Fed launched
Discount Window Direct, an online
portal qualified banks can use to access
the facility.
All eyes will be on these and other
reforms as the Fed (alongside other
regulators) continues to explore ways
to improve its oldest function before
the next crisis. EF

READINGS
“Bank Failures and Contagion: Lender of Last Resort, Liquidity,
and Risk Management.” G30 Working Group on the 2023 Banking
Crisis, January 2024.

Kelly, Steven. “Weekly Fed Report Still Drives Discount Window
Stigma.” Yale School of Management Program on Financial
Stability, April 3, 2024.

Cipriani, Marco, Thomas M. Eisenbach, and Anna Kovner.
“Tracing Bank Runs in Real Time.” Federal Reserve Bank of New
York Staff Reports No. 1104, May 2024.

McLaughlin, Susan. “Lessons for the Discount Window from the
March 2023 Bank Failures.” Yale School of Management Program
on Financial Stability, Sept. 19, 2023.
econ focus

• third quarter • 2024 7

Rethinking Retirement Savings
Many Americans have surprisingly little set aside for retirement.
Why, and what can be done to boost their nest eggs?
BY MATTHEW WELLS

H

ow much is enough to live comfortably in retirement?
Would $30,000 a year be enough? Maybe if you live in
a low-cost area, and the house is paid off. How about
$15,000? It would be a stretch, at best. Yet recent census
numbers indicate millions of Americans over the age of 65
must figure out how to make ends meet on these incomes.
A quarter of seniors, almost 14 million retirees, live on only
$15,000, while a little over half, 29 million retirees, live on
only $30,000 a year. For these Americans, the prospects of a
comfortable retirement appear uncertain.
According to the Fed’s 2022 Survey of Consumer Finances
(SCF), just over 54 percent of families have retirement
accounts such as IRAs, 401(k)s, 403(b)s, or thrift savings
accounts. Among families that do have them, the median value
of those accounts in 2022 was just $86,900 — hardly enough
to last the 20 years of an average retirement. This is especially
true given that the median retiree spends over 10 percent of
his or her income on out-of-pocket medical expenses that
aren’t covered by Medicare or Social Security. Moreover, for
those approaching retirement in the ages 55 to 64, SCF data
indicate that those in the 50th percentile, or the middle of the
pack, have only $10,000 saved in those accounts.
To be sure, there are some bright spots in the picture.
Andrew Biggs, an economist with the American Enterprise
Institute, suggests that Americans are doing well when it
comes to retirement. Among other data points, he notes that
the elderly poverty rate declined from 9.7 percent in 1990 to
6.4 percent in 2018, and that for those contributing to retirement plans, contributions have increased from about
6 percent in 1975 to over 9 percent in 2021.
But even some of the positives carry some negatives. For
8

econ focus

• third quarter • 2024

example, the average IRA/401(k) portfolio balance for those
nearing retirement, among seniors who have such accounts,
increased from $144,000 in 2019 to $204,000 in 2022. That
is certainly good news, but the same SCF survey indicates
these gains were concentrated among higher-income households, while those in the lower 40 percent were worse off.
Further, account balances for households ages 45 to 54 did
not keep pace with inflation, and 35-to-44-year-olds’ household balances declined in nominal terms.
How did it come to be that so many have so little saved for
retirement? And what can be done to help more Americans
save and retire with financial security?

SOCIAL (IN)SECURITY?
Even after including other potential sources of income like
investment accounts, real estate, and businesses, the 2022 SCF
results suggest that half of households will have to rely almost
entirely on Social Security when they enter retirement. But the
average yearly benefit is only about $23,000 — most likely well
below the 75 percent of pre-retirement income financial planners say is necessary to maintain a consistent standard of living
in the post-working years.
The program’s ability even to provide that modest income
is not guaranteed. The 2023 Social Security Trustees Report
identifies a shortfall of $22.4 trillion through 2097, and estimates that it will only be able to pay out 80 percent of scheduled benefits beginning in 2034 unless changes are made to
the program. Potential fixes include adjusting the payroll tax
structure to generate more funding and increasing the age
to qualify for full retirement (currently 67 for those born in

ACTIVE PARTICIPANTS IN RETIREMENT PLANS
(MILLIONS)

1960 or later) or the maximum benefit
A Dramatic Transition in Private Sector Retirement Savings
(currently 70).
Active participants in defined benefit and defined contribution plans, 1975-2021
The shortfall can be traced primarily to demographic shifts. In 1935, U.S.
100
life expectancy was just under 62 years,
and the fertility rate was 2.1 children per
woman. Life expectancy increased steadily
80
over time and was 69.5 by 1957. This was
the peak year of the baby boom, and the
60
fertility rate was 3.5 children per woman.
Today, Americans live to about 79.3 years,
40
and the fertility rate has dropped to 1.8.
Another way to consider this demographic
shift is to look at changes in the old-age
20
dependency ratio, which is defined by the
Organisation for Economic and Co-operation
0
and Development as the number of indi1975
1980
1985
1990
1995 2000 2005 2010
2015 2020
viduals age 65 and over per 100 people of
working age, generally 20 to 64. In 2000,
Defined Benefit
Defined Contribution
the ratio was 20.9; today, it is 32.2. Along
SOURCE: Employee Benefits Security Administration (EBSA 2023). Private Pension Plan Bulletin Historical Tables and Graphs
with increased life expectancy, the Social
1975-2021, Table E7 via Economic Policy Institute
Security Administration cites several factors
contributing to the changing ratio, including increased female labor force participation
only 11 million active participants. In 2021, that number had
and the widespread postponement of family formation, both of
dropped to 12 million participants for DB plans and grew to
which contribute to fewer births.
88 million participants for DC plans. (See chart.)
Why the shift? Employers typically cover the entirety of a
FLAVORS AND TRENDS OF RETIREMENT SAVINGS
defined benefit plan, making them more costly. Defined contribution plans are also more predictable and easier to adminThus, Social Security benefits are unlikely to fully rescue retirister, as employer contributions follow a set formula (for
ees who don’t have enough money salted away. What, then,
example, contributing 3 percent of an employee’s salary),
explains Americans’ apparent lack of retirement savings?
and they do not rely on actuaries to develop cost projecFor decades, both public and private sector employers
tions of benefits to be paid each year. DB plans, on the other
contributed to their workers’ retirement through pensions,
hand, can require employers to make additional contribuknown broadly as defined benefit, or DB, programs. Under
tions in the event of investment losses to meet the benefit
this system, employers pay out a monthly benefit to each
amount they had previously agreed on with their employees.
retired worker, the value of which is determined by the
The inability for some firms to meet those commitments led
worker’s age, length of service, and final salary. Workers typi- to the Employee Retirement Income Security Act (ERISA) of
cally must remain with a firm for a certain number of years
1974 and the Pension Protection Act of 2006, both of which,
to qualify for a pension, but if they do, they then receive that
among other things, mandated stricter funding requirements
benefit for the duration of their retirement.
to ensure employees receive the benefits they were promDefined contribution (DC) plans, on the other hand, are
ised. ERISA also carries additional costs for employers, which
individual accounts funded by the worker’s own contribumay have prompted them to discontinue offering them to new
tions, employer contributions, or both. Examples of DC plans employees.
include profit-sharing plans, 401(k)s, 403(b)s, and employee
From the employees’ perspective, defined contribution
stock ownership plans. Under these programs, there is no
plans also might be preferable because of their portabilguaranteed income; what is available in retirement is whatity. Participants can “roll over” their account balances from
ever has resulted from those contributions, investment gains
a previous employer’s plan into a new one, allowing them
and losses, or company earnings. These programs can be
to continue accumulating benefits wherever they work. DB
sponsored by the employer, or individuals can open their
plans lack this portability in large part because the beneown individual retirement account (IRA). Either way, the
fit formulas they use only account for a worker’s tenure and
workers typically act as their own financial advisors, deciding salary with respect to a specific employer.
how much money to put in, and allocating and distributing
While these changes initially might make putting money
those funds to maximize returns and hedge against the risks
away for retirement appear easier, there is evidence this tranthat come with investing.
sition from DB to DC plans has led to less retirement savings
For the last nearly 50 years, there has been a massive shift
for a significant portion of American workers. The Bureau
away from defined benefit plans toward defined contribuof Labor Statistics reports that of the 66 percent of private
tion options. In 1975, private sector DB plans had 27 million
sector workers with access to a DC plan, only about half
active participants, whereas private sector DC plans had
actively make contributions.
econ focus

• third quarter • 2024 9

Also, as noted above, anyone can open an IRA regardless of
whether their employer sponsors a retirement plan. But data
from the Census Bureau indicate that as of 2014, only
22 percent of workers at businesses without pension plans had
opened one, and under 8 percent were actively contributing.
Managing such plans, and defined contribution plans generally,
can be intimidating for employees, which may explain the poor
participation rates. Additionally, the census data suggested
that with slow earnings growth over time, many workers have
found it challenging to set aside funds for retirement, instead
opting to use the money for current expenses.

WHY SAVINGS HAVE STALLED FOR SO MANY
A 2023 Congressional Budget Office report estimates that
this shift away from DB plans to DC plans accounts for
about 20 percent of the increase in wealth inequality from
1989 to 2019. Data from the SCF indicate that in 1989, the
median household of those approaching retirement had no
money in retirement accounts or DC plans, while those in
the 90th percentile had $161,000. Over time, that difference
has increased dramatically. In 2022, the top 10 percent held
balances over $1 million, while as noted earlier, the median
household in that age group had balances of about $10,000.
The disparities in uptake and active contributions to retirement accounts also extend beyond income levels to ethnic
groups. While nearly 62 percent of White households have
such accounts, a little more than a third of Black households
and just over a quarter of Hispanic households contribute to
retirement accounts, according to the 2022 SCF.
Monique Morrissey is an economist at the Economic
Policy Institute, a progressive think tank. She argues that
with the bulk of retirement account activity occurring in the
upper income brackets, 401(k)s and other similar retirement
accounts have failed to provide most working Americans
with adequate savings for retirement and have instead been
used by more wealthy Americans primarily as tax-advantaged investment opportunities. She notes that the Treasury
Department has estimated contributions to those accounts cost
$138.5 billion in lost revenues in 2021 alone. (Account holders of pretax accounts pay taxes when they withdraw funds in
the future, but those will likely be different than what would
have been paid in current income taxes.) “If we had taken
all the money we had spent on subsidizing 401(k)s, and we
just divvied it up among households and invested in Treasury
bonds with no employer or employee contributions, most
households would be better off,” she argues.
In addition to the changes in the vehicles available for
saving, the rising costs of health care have also eaten away at
Americans’ savings. Medical expenses rise rapidly with age,
as middle-income individuals can expect to pay an average
of $6,000 annually at age 76, and the cost only goes up from
there — as much as $26,000 if they’re fortunate enough to
reach 100, according to a 2023 working paper by economists at
the University of Minnesota, the University of Cambridge, the
Richmond Fed, and the University of Western Ontario. Most of
those costs come from needing to pay more for out-of-pocket
expenses not covered by Medicare, which provides insurance
to Americans ages 65 and older. Those out-of-pocket costs
10

econ focus

• third quarter • 2024

can go toward prescriptions, hospital stays, home health care,
doctor and dental visits, and premiums for any supplemental private insurance and Medicare itself. Medicare also only
fully covers the first 20 days of a nursing home stay, a reasonably common medical need for the elderly. Some of these costs
are covered by Medicaid, but that program is only available to
those with very limited financial resources.
These costs have forced many Americans to make difficult
decisions about how they will allocate already scarce financial resources. According to a 2023 Kaiser Family Foundation
survey, 36 percent of Medicare beneficiaries indicated that
they delayed or went without medical care because of the
costs. Households with Medicare also spend a larger share of
their budgets, unsurprisingly, on health care than households
that do not use Medicare.
The Kaiser Family Foundation also reported that increases
in health insurance premiums for working families outpaced
increases in workers earnings — and the pace of inflation
— between 2003 and 2018, which means less money to put
away for retirement. Rising health care costs also impact
savings through another, more indirect path: Employers
frequently provide health insurance for their employees, and
increasing costs likely means less money available to spend
on wages and pension or retirement plan investment.

POLICY OPTIONS
Morrissey from the Economic Policy Institute sees Social
Security as the best hope for providing retirement security
to working Americans. Because those benefits are a function of both what a worker pays in and increases to the cost
of living, “the return on Social Security contributions is
much more stable and predictable than what you get with
a 401(k),” she argues. But even if Congress addresses the
shortfall and restores long-term solvency, a 2023 report from
Boston College’s Center for Retirement Research suggests
that absent major increases in funding, Social Security will
replace even less of the 75 percent of pre-retirement income
commonly believed to be necessary for maintaining one’s
standard of living into retirement. Passing those increases
is politically controversial, and would come with their own
economic costs, leading policymakers and researchers to look
for alternatives that might increase Americans’ ability to save
for retirement.
Perhaps the most widely considered options involve
expanding access to defined contribution plans, which, as
noted, have tended to produce benefits that disproportionately benefit the wealthy. Much of that expanded access is
taking place at the state level. Nineteen states and two cities
have enacted some form of retirement savings programs for
their private sector workers, the most common of which is
an auto-enrolled Roth IRA. When an employee begins work,
employers deduct between 3 and 5 percent of each paycheck
and place it into an IRA, although the contribution can
increase incrementally over time. For example, California’s
plan starts at 5 percent and an additional 1 percent is added
every year until it reaches 8 percent. Like all other IRAs, they
aren’t tied to an employer, and individuals can elect to opt
out at any time.

In a 2021 working paper, economists at the University of
Oregon, the University of Pennsylvania, Boston College, and
the Urban Institute evaluated the efficacy of OregonSaves,
the state’s auto-IRA plan passed into law in 2015. They
found that between 2018 and 2020, more than 67,700 workers had accumulated more than $51 million in investment
savings, suggesting auto-enrollment mitigates the barrier of
establishing an account. At the same time, the upper bound
of the participation rate among eligible workers was only
62.4 percent — well below the rate in firm-sponsored plans.
Of those opting out of the program, over 30 percent said
that they couldn’t afford to save.
Alicia Munnell, the director of the Center for Retirement
Research at Boston College, argues for requiring employers
to offer plans. “Nothing is going to get better until there’s a
national mandate that says employers have to either provide
a plan or send their employees’ contribution to a public
version of, say, the Thrift Savings Plan [the defined contribution plan for federal government workers].” She also
argues that having access to a plan is more important than
the type of plan. Defined contribution plans may even have
some advantages over defined benefit plans for workers once
they retire. Having stocks and bonds in defined contribution
accounts “may be better than having a fixed nominal benefit that just gets eroded by inflation,” which might happen
under a defined benefit plan.
While legislation at the federal level has yet to be put
forward containing such a mandate, the Retirement Savings
for Americans Act, introduced originally in 2022, would
create a nationwide auto-enrollment program for workers
who do not have access to employer-provided plans modeled
after Uncle Sam’s Thrift Savings Plan. Like other retirement
accounts, it would be portable, and offer a variety of investment options tied to workers’ estimated retirement dates. To
encourage savings, it would also provide certain savers with
a 4 percent match by the government through an income tax
credit.
A similar matching provision was included in the Secure
2.0 Act, which was signed into law in late 2022. Beginning in
2027, the federal government will match up to 50 percent of
a worker’s contribution to his or her retirement plan up to
$2,000, a benefit known as the Saver’s Match. For example,
a worker contributing $2,000 would see the government also
contribute $1,000. The program is meant to encourage saving
among lower- and middle-income Americans; it is available to single tax filers making a maximum annual income of
$20,500 or joint filers making between $41,000 and $71,000
and will adjust annually for inflation.
Some academic research has suggested that Americans
have historically saved for retirement in an optimal way,

meaning they usually accumulated sufficient wealth to maintain their standard of living. A 2006 paper from the Journal
of Political Economy using data from 1992 to 2004 showed
that over 80 percent of households were saving optimally for
retirement during that period, and those who were not were
only minimally below their target. Additionally, in a 2015
working paper, RAND economists Michael Hurd and Susann
Rohwedder looked at consumption capability, or the extent
of one’s ability to consume whatever goods and services
one wants, as a measure of financial wellbeing rather
than income, and found 59 percent of single retirees and
81 percent of couples are prepared for retirement.
While these measures should not be dismissed, many who
are still working feel increasingly uncertain about how they will
get by in their sunset years. According to the Fed’s 2023 Survey
of Household Economics and Decisionmaking, 80 percent of
retirees said they were doing at least OK financially, but only
34 percent of nonretirees thought their retirement savings plan
was on track, down from 40 percent in 2021.
Three long-running trends have been the source of uncertainty in recent decades. First, people are living longer, meaning it is more expensive for society to support lengthy retirements. Second, historically increasing income inequality,
whether from lower wages or replacing DB programs with DC
plans, means many workers have fewer resources set aside.
Third, ongoing increases in the cost of medical care have eaten
up larger portions of savings. The government in recent years
has paid many of these costs through programs like Medicare,
but there are limits to how much of the burden it will carry.
The solutions that have been offered are also controversial. Some object to the prospect of asking people, especially
lower-wage earners, or manual laborers, to work even longer
while wealthy people at the same age can retire. Voluntary
retirement plans can provide opportunities for savings
accumulation, but it is hard for people to save for the future
when living in an increasingly costly present. On the other
hand, public solutions like increasing Social Security and
government-funded programs require either higher taxes,
more debt, or cuts to other government programs, all of
which carry their own costs and organized opposition.
In a recent article, Munnell of Boston College took note
of the 2023 SCF finding that 80 percent of retirees reported
doing okay when it comes to their finances. While this may
be good news, she pointed to another recent finding regarding retirees that might cast a shadow: Their largest regret
(52 percent) when it came to their finances was that they
didn’t save more when they were working. With only
39 percent of today’s workers being able to maintain their
standard of living into retirement, this cohort of retirees is
unlikely to be the last to hold that sentiment. EF

READINGS
“Changes in U.S. Family Finances from 2019 to 2022: Evidence
from the Survey of Consumer Finances.” Federal Reserve Board of
Governors Research and Analysis, October 2023.

McInerney, Melissa, Matthew Rutledge, and Sara Ellen King. “How
Much Does Health Spending Eat Away at Retirement Income?”
Center for Retirement Research at Boston College, August 2022.

Iwry, J. Mark, David John, and William Gale. “Secure 2.0 and
the Past and Future of the U.S. Retirement System.” Retirement
Security Project at the Brookings Institution, Jan. 24, 2024.

Scholz, John Karl, Ananth Seshadri, and Surachai Khitatrakun.
“Are Americans Saving ‘Optimally’ for Retirement?” Journal of
Political Economy, August 2006, vol. 114, no. 4, pp. 607-643.
econ focus

• third quarter • 2024 11

RESEARCH SPOTLIGHT
by brooke hansbrough

Skill Mismatch, Layoffs, and Bouncing Back
Claudia Macaluso. “Skill Remoteness
and Post-Layoff Labor Market
Outcomes.” CESifo Working Paper
No. 10845, December 2023.

B

eing laid off from one’s job often
leads to worse future employment
outcomes. The underlying reasons
for this are unclear, however. Recent
research by Richmond Fed Economist
Claudia Macaluso has found that
mismatch between a laid-off worker’s
skills and the skills involved in other
local jobs plays a significant role. She
created a novel measurement of “local
skill remoteness” and used it to compare
the effects of layoffs from jobs with
varying levels of this skill remoteness on
a worker’s wages, future employment,
and migration.
Macaluso defined skill remoteness as “the degree of dissimilarity
between the skill profiles of a worker’s job and all other jobs in a local
labor market.” This calculation has two
crucial components: the differences in
the skill content of jobs (not of workers),
and the differences in job availability
across geographic locations. For example, an economist would likely have
higher skill remoteness than an office
manager, since more jobs involve social
and administrative skills than advanced
economics. Two economists in different locations would also have different
levels of skill remoteness: An economist
on Wall Street would have more skills
in common with the other jobs in the
area than an economist in rural Iowa.
To account for all of this, Macaluso
created measurements of the
“distance” between a particular occupation’s skills and that of every other
occupation, and then weighted the
distances by the prevalence of that
occupation in a certain city each
year. This data-heavy task hinged
on a survey called the Occupational
Information Network, or O*NET,
which asks workers and occupation

12

econ focus

• third quarter • 2024

experts to quantify the skills they use
in their occupations, among other
characteristics.
Armed with this empirical
measurement of skill dissimilarity
across geography and time, Macaluso
set about investigating its effects.

Skill mismatch matters,
Macaluso suggested, and
can play a significant negative
role in labor market outcomes,
especially for displaced
workers.
Skill remoteness is not inherently a
bad thing; niche skills could be valuable and well-rewarded in the labor
market. But they could also make it
more difficult for skill-remote workers to find wages on par with their
skills in their local areas. Macaluso
found more evidence of the latter in
her investigation of post-displacement
behavior for those who lost their jobs
through no fault of their own (plant
closure or layoff ). Using the National
Longitudinal Survey of Youth 1979,
she tracked workers over their careers
and confirmed the literature’s finding of a large and persistent earnings
drop after displacement; on average,
displaced workers earned only about
60 percent of pre-displacement earnings four years later. This is true even
when accounting for other factors that
could affect displacement and earnings, such as location, occupation,
date, and local unemployment rate.
The main results of Macaluso’s
investigation lie in comparing skill-remote (above median skill remoteness) to skill-central (below median)
displacements. She discovered that
there was no association between
pre-layoff earnings and skill remoteness, but there was a strong negative

correlation post-layoff. Earnings in
the month of layoff were almost $500
per month lower for skill-remote
workers than skill-central, and this
persisted at around $200 per month
less even four years after the displacement. Over the course of these four
years, this difference added up to
over $10,000, a substantial loss. She
also found that workers displaced
from skill-remote jobs had a lower
probability of working jobs with
similar skill profiles in the future.
Additionally, workers in locally
skill-remote jobs were more likely to
migrate to cities with a lower local
skill remoteness (our displaced Iowa
economists tended to move to places
like New York City). Not only did they
change cities, but workers who lost a
skill-remote job were also 11 percent
more likely to change occupation than
those who lost a skill-central job, and
they went through more substantial
skill portfolio changes.
Macaluso found the business cycle
was important to this, as well; earnings
losses following displacement are more
severe in recessions, and the fraction of
destroyed jobs that are skill-remote is
higher in recessions (60.3 percent) than
in booms (46.6 percent). This suggests
that labor policies that target individuals displaced from skill-remote jobs
could ameliorate some post-layoff hardships, especially in recessions.
The magnitude of these earnings losses and the other disruptions
associated with skill remoteness are
significant. Macaluso stressed that
skill remoteness is a quality of jobs,
not workers: Skills can be taught,
and workers can potentially improve
their post-layoff prospects by investing in skills valued in their local area.
Skill mismatch matters, Macaluso
suggested, and can play a significant negative role in labor market
outcomes, especially for displaced
workers. EF

POLICY UPDATE
b y s a m l o u i s tay l o r

Independence, If You Can Keep It

H

istorically, Congress has tended
to take an acute interest in examining the structure of the Federal
Reserve whenever there is economic
turmoil. The economic swing during
the COVID-19 pandemic and the
current period of elevated inflation
are no different. In response, lawmakers and policy influencers have voiced
concerns about the Fed’s ability to
promote an equitable economic recovery as well as its ability to manage
inflation. These proposals have
spanned the political spectrum, including expanding the Fed’s monetary
policy mandate into new areas, bringing monetary policy decision-making
under additional oversight, and replacing the current structure of the Fed
entirely.
At the same time, other lawmakers
have supported monetary policy independence. “Given [the FOMC’s] charge,
their independence is critical to doing
it in an unbiased, nonpolitical way,”
Sen. Kevin Cramer, R-N.D., told the
Wall Street Journal in April.
In the modern era, the Fed has
largely been granted independence to
conduct monetary policy without direct
interference from elected leaders. What
exactly does it mean for a central bank,
created by Congress, to be “independent,” and how is the Fed in its current
form accountable to elected leaders and
the public at large?
Congress has mandated that the
Fed pursue two objectives in conducting monetary policy: promote the goals
of maximum employment and stable
prices. Congress has given the Federal
Open Market Committee (FOMC) wide
latitude in how it pursues those goals.
This is what economists call “instrument independence.” Political interference in these instruments, many
fear, would lead central bankers to be
more responsive to short-term demands
of politics and could harm the country’s long-term economic stability.

Economists and historians have argued
that insulating monetary policy from
political direction results in better longterm outcomes in the form of lower
rates of inflation, among other benefits.
This independence has evolved over
time. From the Fed’s inception in 1913
until 1951, the Fed was much more
closely tied to the executive branch.
During both world wars, the Fed set

Congress has given the Federal
Open Market Committee wide
latitude in how it pursues its
dual mandate. This is what
economists call “instrument
independence.” Economists
and historians have argued that
insulating monetary policy from
political direction results in better
long-term outcomes in the form
of lower rates of inflation, among
other benefits.
explicit expectations that it would help
to finance federal spending and support
government bond prices. During the
Great Depression, monetary policy was
largely dictated by Congress and the
Roosevelt administration, giving the
Fed little leeway to deviate from those
instructions. This arrangement continued in the immediate postwar era until
concerns over rising inflation led to
confrontation in 1951.
In what is commonly referred to
as the Treasury-Fed Accord of 1951,
the Fed and then assistant secretary of the Treasury William Martin,
himself a future Fed chair, agreed
that the Fed could conduct monetary policy without approval from
the executive branch. Though political interference in the Fed would
continue to fluctuate in future years,

this accord marked the modern era of
operational independence for monetary policy.
Together with this independence,
the Fed is regularly subject to oversight by Congress for its activities
and decision-making. The Board of
Governors prepares a report on the
FOMC’s monetary policy actions twice
a year, which is accompanied by testimony from the chair of the Board of
Governors to both houses of Congress.
Additionally, since 2010, the vice
chair for supervision has testified to
Congress twice a year on banking
conditions and the Fed’s regulatory
actions. The Fed is regularly audited
by the Government Accountability
Office (GAO) and the Office of the
Inspector General, as well as independent outside auditors. Those reports
are publicly available. (Congress did
create a limited exemption from GAO
audits for monetary policy deliberations and actions to avert political
interference.)
To provide additional transparency into its decision-making, the
FOMC issues public statements on its
rate decisions and releases its meeting minutes and transcripts. FOMC
members also regularly make public
comments and take questions from
the media. Most notable is the chair’s
press conference immediately after
each rate-setting meeting.
Proponents of central bank independence concede that this authority must be used responsibly to pursue
the goals set out by Congress. “As we
move along the path of reform .… it
is crucial that we maintain the ability of central banks to make monetary policy independently of short-term
political influence,” argued then-Fed
Chair Ben Bernanke in a 2010 speech.
“In exchange for this independence,
central banks must meet their responsibilities for transparency and accountability.” EF
econ focus

• third quarter • 2024 13

i m ag e : court e sy of ch e sterfie l d food ban k

B Y M AT T H E W
WELLS

Food Banks: Lifelines to
Those in Need
For millions, food banks fill a crucial gap. How do they do
it and just how big is the need they address?

I

t’s a Friday night and food distribution at the Chesterfield Food Bank Outreach Center, pictured
above, is in full swing. Inside the cavernous warehouse in suburban Richmond, across the street
from a veterinary clinic and an auto shop, staff members keep everyone on task while the latest pop
hits keep the beat. Scores of volunteers — retirees, religious and scout groups, and families — work
the different sections of fresh fruits and vegetables, frozen meats, canned and dry goods, and more.
Many sing along as they fill the shopping carts of other volunteers cycling through a canyon of pallets and
boxes of food, making their way outside. There, another volunteer will direct them to one of the lanes with a
waiting car, where they’ll unload a week’s worth of groceries, then return inside to load up and do it again.
This scene repeats the first and third Friday of each month at the food bank’s main facility, as well
as every second and fourth Monday at a local church, and every Saturday at a nearby school or similar venue. Nick Jenkins, the food bank’s community outreach director, says that about 20,000 to 25,000
visitors use their services each month, representing almost 5,000 families in the surrounding area. “I
think most people don’t associate Chesterfield County or America with hunger, but I think it’s important to understand what food insecurity here is,” he says. “It’s people that are maybe not going seven
days without food, but they don’t have the security to know that when they go home, they have food in
their pantry or the means to purchase food.”
The need is striking: Food distribution starts at 4:30 p.m., and by the end of the night around 9 p.m.,
somewhere between 400 and 600 shopping carts of food, usually one per car, will make their way to
the homes of some of the community’s most vulnerable residents.
Food banks and other charitable food organizations like this one exist to meet the nutritional needs
of community members throughout the Fifth District and across the country. Doing so requires an

14

econ focus

• third quarter • 2024

array of partnerships to acquire food, staff and operational
funds, an understanding of communities’ needs and challenges, and logistical creativity to ensure food gets to those
who need it.
WHAT DO FOOD BANKS DO?
The core mission of food banks is to provide food to those in
need. General distribution efforts like the one in Chesterfield
are common, but theirs is just one of the many approaches.
Other programs directly support food access for children. For
example, the Capital Area Food Bank in Washington, D.C.,
participates in the U.S. Department of Agriculture’s (USDA)
Summer Food Service Program, hosting 22 sites where children under the age of 18 in and around the capital can receive
a free nutritious lunch. Seniors and homebound residents also
benefit from specialized programs, including the well-known
Meals on Wheels, which is run out of food banks such as Feed
More in central Virginia.
What is a food bank? There is flexibility in the term, but,
generally, food banks are large, regional organizations that
store and distribute food to more local programs like food
pantries. Food pantries then distribute the food to those in
need. Some food banks serve vast geographic areas, such as
Feed More, which has a coverage area of 29 counties and
five cities from Virginia’s southern border to the Northern
Neck. Others, like the Capital Area Food Bank, serve more
densely populated urban areas. Chesterfield Food Bank
operates as both a food bank and a food pantry, and it is
a part of the Feed More distribution network. Both food
banks and pantries supply prepackaged food, as well as
fresh produce and frozen meat, which are taken home and
prepared. They differ from soup kitchens, which cook and
serve food at a set time.
According to Feeding America, a nationwide network
of 200 food banks and 60,000 partner food pantries, food
banks acquire their food through donations, purchasing,
and federal programs. First, community partners like local
grocery stores, restaurants and bakeries, and small businesses donate their overages, the surplus food beyond what
is needed to stock their shelves or serve the day’s customers. Food drives also allow those within the local community to help out, as individuals, community organizations,
and businesses can all collect, then donate, food that they
have gathered. Local farmers also play a key role in supplying healthy and fresh produce, as they will oftentimes
donate “perfectly imperfect” food — produce that is fine
but might not meet the aesthetic expectations for retail
sale. For example, Feeding the Carolinas, a network of the
10 regional food banks across North and South Carolina,
established Farm to Food Bank, a program pairing the food
banks with over 50 Carolina farmers who make available
over 35 different varieties of produce for families in need.
Second, food banks and pantries purchase food — which,
when sufficiently scaled, can be done at much lower prices
than when individual consumers visit the grocery store.
Doing so allows a food bank to tailor its offerings to meet
the specific cultural or medical needs of its clients. Third,
food banks also receive food through government programs

like the Commodity Supplemental Food Program, which
provides necessary items to senior citizens, such as milk,
fruits and vegetables, cereal, and cheese.
Food banks and pantries could not operate without volunteers. Community members might volunteer for a range of
reasons: Many belong to scouting or religious groups, some
are looking to fulfill service hours for school, and others are
families or retirees wanting to give back to their communities. Jenkins says volunteers contribute about 60,000
hours per year at the Chesterfield Food Bank, with between
30 and 50 people working shifts daily at the main facility
doing data collection and reporting or working in the warehouse and pantry. An additional 75 to 150 usually work at
distribution events, and others help with outreach, staffing
programs to make sure community members in need know
how to access their services.
These outreach efforts — which include visiting local
motels and homeless encampments, purchasing advertising space on local television, and ensuring Chesterfield Food
Bank comes up whenever anyone in the area searches “food
insecurity” online — are crucial. “We constantly assume that
no one knows us,” says Jenkins. “So we share our name in
a hope that it gets in front of people who need help.” Some
organizations maintain eligibility requirements (for example,
that recipients must be under 200 percent of the poverty line
to receive food or must be referred by another social service
organization), while others encourage anyone in need,
regardless of the circumstances, to use their services.
Like many charitable organizations, food banks and
pantries rely heavily on philanthropy to fund their operations. For example, the Summer Food Service Program
administered by the Capital Area Food Bank is sponsored by
several partners, including grocery store chain Harris Teeter,
as well as DC United, the city’s professional men’s soccer
team. Many charitable food websites, from Feeding America
at the national level to state-level networks like Feeding the
Carolinas and local facilities like the Chesterfield Food Bank,
make it as easy as possible for individuals to donate money,
and they frequently tell donors exactly what their money will
do. In Chesterfield, more than half of the food bank’s funding
comes from donors in the community who give a recurring
donation of $25 to $50 a month. The remainder comes from
grants from local governments, businesses and civic organizations, as well as national charitable foundations such as the
Walmart Foundation.
THE PROBLEM OF FOOD INSECURITY
People turn to food banks in varying circumstances. In some
cases, a job loss or other disruption can lead households to
seek food assistance to cover short-term needs, perhaps just
for a month or two. In other situations, such as for households
on fixed incomes, reliance on charitable food may be built into
their standard approach for meeting basic food needs.
Feeding America reported that in 2022 alone, 49 million
people relied on free food assistance, and that its partners
distributed over 5.3 billion meals. The Urban Institute’s Wellbeing and Basic Needs survey found that the following year,
in 2023, one in six, or 16.6 percent of adults in the United
econ focus

• third quarter • 2024 15

States, reported receiving charitable food; in 2019, prior to the
COVID-19 pandemic, that number was closer to one in eight.
These numbers indicate how many people rely on charitable food organizations like food banks, but do they accurately reflect the real level of need? Maybe not: A deeper
look into the data from the 2023 Urban Institute survey
reveals that those who stated they used food bank services
were only about 40 percent of the overall number of those
who reported experiencing food insecurity.
The USDA defines food insecurity as “the limited or uncertain availability of nutritionally adequate and safe foods, or
limited or uncertain ability to acquire acceptable foods in
socially acceptable ways.” Households are classified as food
insecure if “they had difficulty providing enough food for all
members at some time in the past year because there wasn’t
enough money for food.” The National Institute for Minority
Health and Health Disparities notes that this insecurity puts
people at risk of other dangerous health conditions such as
diabetes, obesity, heart disease, and mental health disorders.
It’s also important to note that food insecurity is distinct
from hunger, which the USDA defines as a physiological
condition that “results in discomfort, illness, weakness or
pain that goes beyond the usual uneasy sensation.”
At the national level, a 2023 USDA report found that 44.2
million people experienced food insecurity in 2022, which
was 10.4 million more than in 2021. More than 13 million
children experienced food insecurity in 2022, as well — a
nearly 45 percent increase over the prior year. Food insecurity also tends to be concentrated in poorer and minority
households; the Urban Institute survey found that for adults
with family incomes at 200 percent of the poverty line
and below, 52.2 percent reported experiencing food insecurity in 2023, compared to 46.6 in 2022. Also, nearly 39
percent of Hispanic adults and about 35 percent of Black
adults reported experiencing food insecurity, both of which
were increases over previous years. (Other racial and ethnic
groups did not experience any increases over the prior
year.) Finally, adults who lived with children, identified as
LGBT, or were low-income renters (as opposed to homeowners) were more likely to report experiencing food insecurity at some point in 2023.
Food insecurity exists in every county in every state.
According to the Food Research and Action Center, a
hunger research and policy center, West Virginia has one
of the highest food insecurity rates in the country at 15
percent, or almost 270,000 residents. Approximately 74,000
children in the state face food insecurity. North Carolina
isn’t far behind, according to their estimates: Nearly 1.5
million North Carolinians (14 percent of the population)
experience food insecurity, and 448,000 of them are children. Almost 750,000 (12.5 percent) Marylanders — 221,000
of whom are children — are food insecure, as are 679,000
South Carolinians, or 12.8 percent of the population.
197,000 of those residents are children. In Virginia, just
under 964,000 residents (11.1 percent) have experienced
food insecurity, and of those, over 252,000 were children.
Finally, in Washington, D.C., 11 percent of the population,
about 74,000 people, experience food insecurity, and 18,000
of them are children.
16

econ focus

• third quarter • 2024

FEEDING PEOPLE ISN’T EASY
An eye-catching data point from the Urban Institute survey:
Only 37 percent of food insecure respondents felt comfortable using charitable food. Kassandra Martinchek, who
studies food insecurity at the Urban Institute, suggests that
the stigma attached to needing free food may keep many
away. Turning to free groceries and meals oftentimes brings
up questions of whether people feel they are deserving or
should access these resources that are available to them.
A separate Urban Institute study of Arlington County,
Va., asked individuals reporting food insecurity who did not
take charitable food to explain why. “I think I’m like everybody else,” said one respondent. “We tend to be proud. In
fact, that would be probably my last resort because I just
believe in, I guess, pulling yourself up by your bootstraps.
I wouldn’t want to do that [accept charity], but I do know
where the resources are if I had to.”
While some may know how to access charitable food, if
necessary, others acknowledge they lack that information;
under half, about 42 percent of those identifying themselves as food insecure, were aware of local charitable food
resources such as food banks or pantries. Beyond the sense
of pride keeping some from exploring the options available to them, this lack of awareness may also stem from a
number of other factors, including not knowing where to
look, unreliable or limited internet capability, and, for those
who do not speak English, information not being available
in their language.
Not knowing about the charitable food resources that
are available is just one dimension of the larger problem
of access. Finding transportation can be difficult for many
such as the elderly and disabled, and even if a charitable
food site is within a reasonable distance and can be reached
via public transportation, operating times may not match up
with the schedules of those in need, particularly if they are
working. Arlington County, for example, had 56 total charitable food sites as of 2022. Forty-eight of those were open
year-round, but only 21 had no eligibility requirements.
Of those, only 10 made food available to residents at least
once a week. Finally, of those 10, only six were open during
weekends or evenings.
For food insecure households with adults working during
regular business hours, these limitations pose a serious risk
to their ability to get enough food. To combat this, charitable food organizations have developed several approaches
intended to make it easier to get food to the people who
need it. The Food Bank of Central and Eastern North
Carolina, for example, has experimented with a graband-go model, where clients who aren’t able to pick up food
during working hours can do so at a time that works for
them. “If someone’s working eight to six, we need to make
sure that we get food resources to them at a time and place
that they can access them or else we’re not doing our job,”
says Marlowe Foster, the food bank’s former senior vice
president for development and business strategy.
Others have leveraged emerging technologies to ensure
people get the food they need. In San Francisco, DoorDash,
the online food delivery platform, partnered with local food

organizations in 2018 to develop Project DASH, a home
delivery service that, according to its clients, helped them
save money, cut down on trips to the grocery store, save
money on transportation, and extend the duration of their
other public benefits. The program partnered with more
than 300 nonprofits by the end of 2022, and food banks and
United Way affiliates around the country have now adopted
the approach. Still others, such as Feed More in Virginia,
partner with local schools to send home weekend backpacks full of food and host farmer’s market-style distributions, which oftentimes will have music, games, and other
community resources.
While Project DASH has delivered about 15 million
meals to rural areas since its creation in 2018, putting food
into the hands of people experiencing food insecurity in
those regions has been a persistent challenge, as they may
be hours away from a food bank or grocery store. Feeding
America notes while 63 percent of all counties in the United
States are rural, 87 percent of counties with the highest
rates of food insecurity are rural. A 2017 paper by researchers at Feeding America and the University of Illinois
showed that charitable food providers are heeding this
reality, as there is twice as much coverage in terms of the
number of food providers in rural areas compared to urban
ones. There are also more distribution days per person in
rural areas and more food given out per person. This is
due in large part to the significant presence of permanent
food pantries, as well as mobile pantries, where food can go
straight from a delivery truck to families in need.
PANDEMIC-ERA BENEFITS AND INFLATION
The COVID-19 pandemic had a dramatic effect on
food insecurity and the work of food banks. Prior to
the pandemic in 2019, 4.4 percent of American households reported they used a food pantry that year, but
that number jumped to 6.7 percent in 2020, the year
the pandemic began. With job losses mounting, the
federal government increased the monthly payment
(known as emergency allocations) to recipients under the
Supplemental Nutrition Assistance Program (SNAP), the
largest federal anti-hunger program. It also relaxed eligibility requirements, which, until then, included a work
requirement. The number of individuals participating in

the program went from about 37 million in February 2020
to 43 million later in the year.
But in the first half of 2023, the work requirement was
reestablished, and states eliminated the emergency allocations. While the number of recipients remained above 40
million, the average SNAP recipient lost between $91.50
and $112.18 per month. This reduction was hard for low-income families, who, like everyone else, were also experiencing the reduced value of their dollars at the grocery
store because of the worst inflationary episode in 40 years.
Sixty-five percent of Feeding America’s food bank partners
reported an increase in the number of people seeking assistance between February and March of that year, suggesting
the lower SNAP amount coupled with crippling increases in
food prices was leading to more widespread need.
Wages in the aggregate are now outpacing inflation, but
many lower-income households who qualify for SNAP are
still struggling to get by, as the benefit doesn’t last them
through the month. As these households turn to charitable
food to fill those gaps, food banks are also experiencing the
effects. Martinchek of the Urban Institute notes while food
banks and pantries received increased funding and donations in the immediate wake of the pandemic, these funds
have expired, even though demand exceeds pre-pandemic
levels. She suggests “it could become increasingly challenging for these food banks to be able to fully meet the needs
their communities have, especially when are also facing
inflationary pressures when they’re purchasing food for
their communities.”
In addition to providing food, many charitable food organizations also make available wraparound services that
foster self-sustainability, including job training programs
and mental health or substance abuse treatment. Foster, the
former head of development for the Food Bank of Central
and Eastern North Carolina, says that the process of
connecting people to these services is a “slow build” requiring trust and communication.
But the children experiencing food insecurity don’t
understand any of that, which is why in Chesterfield, cars
with a child in them will receive a cake or pack of cupcakes.
“A kid sitting in a line for an hour and a half at the food
bank waiting for groceries is not really understanding
everything that’s going on,” says Jenkins. “But at the end of
it, if they get a cake, it makes them happy.” EF

READINGS
Gundersen, Craig, Adam Dewey, Monica Hake, Emily Engelhard,
and Amy Crumbaugh. “Food Insecurity across the Rural-Urban
Divide: Are Counties in Need Being Reached by Charitable Food
Assistance?” Annals of the American Academy of Political and Social
Science, July 2017, vol. 672, pp. 217-237.
Gupta, Poonam, Elaine Waxman, Michael Karpman, Baris Tezel,
and Dulce Gonzalez. “Food Insecurity Increased for the Second
Straight Year in 2023,” Income and Benefits Policy Center, Urban
Institute, April 2024.

Martinchek, Kassandra, Theresa Anderson, Poonam Gupta,
Fernando Hernandez-Lepe, Alena Stern, and Amy Rogin.
“Improving Food Security and Access in Arlington County,
Virginia: Mixed-Methods Analyses and Recommendations.”
Income and Benefits Policy Center Research Report, Urban
Institute, March 2022, revised April 2022.
Norris, Stephanie. “Households Confront the End of Pandemic-Era
Assistance Programs.” Regional Matters, Federal Reserve Bank of
Richmond, July 13, 2023.

econ focus

• third quarter • 2024 17

ECONOMIC HISTORY
by julian kikuchi

Milton Friedman, Dissenter

A

t an event in 2002 in honor of
Milton Friedman’s 90th birthday,
then-Fed Chair Ben Bernanke
offered him an olive branch of sorts
on behalf of the Fed. “Regarding the
Great Depression. You’re right, we did
it,” Bernanke conceded. “We’re very
sorry. But thanks to you, we won’t do
it again.”
Bernanke’s comment was an allusion to the 1963 book A Monetary
History of the United States 1867-1960
by Friedman and economist Anna
Schwartz, in which they argued that
monetary policy led by the Fed had an
enormous influence on the recessionary
periods of the U.S. economy, including the Great Depression. That view,
although contradictory to the general
belief of the time that money had little
role in economic fluctuations, became
increasingly important and influenced
policy responses of the Fed during the
2008 financial crisis.
Throughout his career, Friedman
was an advocate for monetarism and
free markets. He believed that a stable
monetary framework, characterized
by steady growth in the money supply,
was essential for fostering economic
stability and prosperity. Moreover, he
considered free markets to be the best
way to allocate resources and deliver
economic prosperity, and he thought
they went hand in hand with individual
freedom. These ideas were long out of
favor in academic circles but turned out
to be another area in which, over time,
Friedman would see much of mainstream thought move toward his views.
STARTING A CAREER IN A TIME
OF CRISIS
A talented student, Friedman graduated from high school before turning
16 and pursued his college education at
18

econ focus

• third quarter • 2024

Rutgers University. As
he was mathematically
inclined, he planned
at first to major in that
subject. Fate intervened in the form of
Arthur Burns, a Ph.D.
student at Rutgers,
who introduced him
to the field of economics through his course
titled “Business
Cycle.” (As it would
turn out, Burns later
became chair of the
Federal Reserve Board
of Governors and
presided over the high
inflation of the 1970s.)
Burns introduced
Friedman to important ideas in economics. First, he introduced him to the
theory of the business cycle. Second,
he introduced him to
Economist Milton Friedman
Alfred Marshall’s ideas,
known as marginalism, which describe
of Chicago. Friedman was introduced
how the marginal utility individuals
to Knight’s work by an instructor at
obtain from consuming an additional
Rutgers who had Knight as his doctoral
unit of service or goods influences their
advisor. In a course on insurance, the
economic decisions. Burns, impressed
instructor used Knight’s book Risk,
with the young student, asked him to
Uncertainty and Profit, in which Knight
proofread his dissertation. Through his
defined risk and uncertainty as sepainteractions with Burns, not only did
rate concepts that can be measured
Friedman learn the practice of scholar(risk) or cannot be measured (uncership, the bond between the two became
tainty) and used them to discuss more
increasingly strong, to the point where
fundamental questions in economics,
Friedman regarded him as a “surrogate
such as, “Why do profits exist?”
father” — possibly influenced by the fact
Upon graduating in 1932, Friedman
that Friedman had lost his father just
decided to pursue graduate education
before entering Rutgers.
at the University of Chicago in economWhile at Rutgers, Friedman was
ics instead of his alternative option
introduced to another important idea
of studying mathematics at Brown
in economics through a book by Frank
University. The University of Chicago
Knight, a professor at the University
economics faculty at the time was not

i m age : t h e f r i e dm an fo u n dat i o n fo r e d u c ati o n a l c h o i ce

The architect of modern monetarism was also an energetic public intellectual

known for its laissez-faire perspective,
as it would be in later years. He was
also exposed to political questions of
the time. The Great Depression, which
was then underway, was blamed by
many on capitalism itself; there were
protests advocating for different social
forms such as communism and social
democracy while fascism was spreading in Europe, and militaristic Japan
was flexing its muscles.
Of that period, economist Allan
Meltzer later wrote, “The dominant
view then was that capitalism had
failed; the future was some form of
socialism, and the only issue was how
extensive it should be. Keynes wanted
free markets for consumer goods but
state planning and direction of investment.” Many economists associated
market competition with waste, as
in the case of multiple milk companies delivering milk to the same block;
there was less attention to the benefits
competitions can offer to consumers.
After learning that his advisor at
Chicago would be away, Friedman
decided to spend his second year at
Columbia University. Columbia was
known for a still more interventionist approach to economics. Economists
at Columbia generally thought that the
economy would not naturally reach
an equilibrium, and that having wellplanned active policies by the government was important for addressing
economic issues. Attending seminars
at Columbia, Friedman was exposed
to different ideas on how to address
the depression the country was facing
at the time. He also added another
dimension to his training by taking
mathematical economics classes from
Harold Hotelling. This was an important step for Friedman since economics research was becoming increasingly
mathematical.
As a result of his mixed education
at Rutgers, Chicago, and Columbia,
by the end of his second year of graduate school, Friedman was exposed
to a broad range of approaches to
economics, including neoclassical price
theory, the quantity theory of money,

institutional economics, and mathematical economics. His exposure to different strands of economics as well as his
mathematical maturity helped form his
foundation. As historian Jennifer Burns
put it in her 2023 biography, Milton
Friedman: The Last Conservative, “The
choices before Friedman were clear;
he was in an ideal position to chart his
own path as a scholar.”
Friedman’s early career was in statistics. In 1935, Friedman, now in his
early 20s, got a job as a statistician at
the National Resources Committee, a
New Deal agency in Washington, D.C.
Foreshadowing his later work related
to inflation, he was assigned to develop
methods to calculate weights for the
consumer price index. He then worked
for the Treasury in its Division of Tax
Research from 1941 until 1943. At that
point, Friedman moved to Columbia’s
Statistical Research Group, which was
directed by Hotelling; there, Friedman
assisted in the use of statistics in
war-related projects.
THE RISE OF MONETARISM
After his work as a statistician,
Friedman took a position at the
University of Chicago in 1946 to teach
a course in price theory. In the decades
that followed, Friedman made major
contributions in macroeconomics, while
a group of Chicago microeconomists,
such as George Stigler and Ronald
Coase, challenged then-dominant views
favoring government intervention. What
emerged from their combined work
was a “Chicago School” of thought that
highlighted what its members viewed
as the importance of individual freedom
and limited government interventions
for economic prosperity.
While Friedman’s most famous
contribution, monetarism, was set out
in his 1963 book A Monetary History
of the United States 1867-1960 with
Schwartz, it came into its own in the
1970s when it gained influence with
policymakers. The book analyzed
major economic fluctuations the United
States experienced from 1867 to 1960

and described the role monetary policy
played in these events. Robert Hetzel,
a doctoral student of Friedman’s and
a former Richmond Fed economist,
observed, “A Monetary History was
one of the most influential books of
the 20th century because of the way it
radically altered views of the cause of
the Depression.” Both Fed Chair Paul
Volcker and British Prime Minister
Margaret Thatcher used elements of
Friedman’s work to tame the inflation
each of their countries was facing in
the 1970s to early 1980s.
Monetarism asserts that in the long
run, the money supply determines the
price level — or, as Friedman put it in
1970, “Inflation is always and everywhere a monetary phenomenon.” (He
later clarified that he was referring to
persistent inflation.) Thus, in his view,
central banks’ objective of price stability
would be best achieved by targeting the
long-run growth rate of money supply.
Underlying monetarism is a concept
called the quantity theory of money,
which comes from a simple accounting
identity: MV = PQ, where M represents
money supply, V represents velocity
(how often a dollar changes hands), P
represents price level, and Q represents
quantity of goods and services bought
and sold. According to monetarists,
V stays relatively constant over time.
Thus, changing the money supply would
inevitably — mathematically — change
the price level. Friedman had been
exposed to these ideas at Chicago by
Knight and by economist Henry Simons.
The Keynesians, so named for the
British economist John Maynard
Keynes, had a different view of
economic fluctuations and inflation. They did not believe that the
kind of monetary policy advocated by
Friedman would naturally lead to desirable economic outcomes. Rather, they
argued that achieving full employment
required the government to use fiscal
policy to influence aggregate demand.
These differences in views regarding how macroeconomic equilibrium
is achieved and the role of monetary policy had implications for how
econ focus

• third quarter • 2024 19

20

econ focus

• third quarter • 2024

Milton Friedman (right) and producer Robert Chitester during the production of the 1980 television series “Free
to Choose.”

argued that consumption depends on
permanent income. He later recalled,
“The catalyst in combining my earlier
consumption work with the income
analysis in professional incomes into
the permanent income hypothesis was
a series of fireside conversations at our
summer cottage in New Hampshire
with my wife and two of our friends,
Dorothy S. Brady and Margaret Reid,
all of whom were at the time working
on consumption.”
The permanent income hypothesis had fiscal policy implications and
clashed with the ideas presented by
Keynes. As in the controversy over
government interventions to maintain macroeconomic equilibrium,
Keynes’ theory lent support to government interventions to help economies
escape from economic downturns. His
absolute income hypothesis suggested
that to avoid recessions, governments
should transfer money to citizens or
increase government spending to boost
the income of the recipients to encourage them to spend or invest the money
received. The absolute income hypothesis implied that people will increase
spending and investment as their
income rises, whether temporarily or

not. In contrast, Friedman’s permanent
income hypothesis posits that an individual’s consumption is driven by his
or her anticipated permanent income;
under that view, if governments transfer money or increase government
spending to raise the income level of
their citizens, and if the citizens regard
the increase in income as temporary,
then the effect on their spending — and
thus on the economy — will be modest.
Today, state-of-the-art macroeconomic
models incorporate the permanent
income hypothesis for some consumers, while allowing for the possibility
that borrowing constraints force some
households to consume based on their
current income.
FRIEDMAN’S OUTREACH TO
THE PUBLIC
Beyond his research, Friedman was a
highly active and visible public intellectual. He engaged extensively with the
public through books and articles and
participated in debates and forums. His
most notable engagements included his
1962 book Capitalism and Freedom and
his 1980 TV series “Free to Choose”
and its accompanying book.

i m age : co u rt e sy o f f r eeto c h o os e n e t wo r k / w i k im e d i a com mo n s

E C ONOMIC HIS TORY

monetarists and Keynesians looked at
the Great Depression. The Keynesians
interpreted the depression as an aggregate demand shortfall that was best
remedied by fiscal policy. Their influence stimulated the “fiscal revolution”
in America, which changed the view of
the government budget from a means
to support small but necessary government functions to a tool for stabilizing
the economy. Friedman and Schwartz,
on the other hand, viewed the Great
Depression as a monetary policy failure in which the Fed failed to provide
banks with necessary cash to avoid
bank failures from bank runs.
Another area where Friedman made
a notable contribution is the permanent income hypothesis. Drawing upon
initial work by economist Dorothy
Brady and by his wife, Rose Friedman,
he developed a theory and provided
supporting empirical evidence that individuals’ consumption depends on their
long-term income prospects — that is,
their “permanent income” — rather
than simply their current incomes.
Keynes had earlier argued that as
people earned money, they increased
consumption, though not by as much
as the increase in income. Economists
framed the problem mathematically and
called the relationship “the consumption function.” Keynes asserted that
the division between consumption and
savings is determined by the disposable
income of the person and coined the
term “absolute income hypothesis.”
Rose Friedman and Brady challenged
Keynes’ absolute income hypothesis. In
a paper titled “Savings and the Income
Distribution,” they argued that household saving and consumption rates
depended on their relative income but
not the absolute income within their
neighborhood, and this was known
as the “relative income hypothesis.”
Margaret Reid, another economist, also
contributed to their research in this
area.
Building on this work, Friedman
modeled consumption as a function
of permanent income and transitory
income and through his analysis, he

Through Capitalism and Freedom,
Friedman advocated the idea that a
rising standard of living is a result of the
free market. He made recommendations
in the book on a wide range of policy
issues in areas such as taxation, education, licensing, and exchange rates.
Friedman expanded his reach
further when he started writing a
regular column for Newsweek magazine in 1966. In 1970, in a New York
Times article, he argued against broad
interpretations of corporate responsibility, holding that the main responsibility of businesses is to generate as
much profit as possible for their shareholders. He contended that the government is responsible for serving the
social interests by taxing the population and providing public infrastructure. Corporations, in contrast, are
responsible for doing things that serve
the best interest of the company, not of
society as a whole.
The “Free to Choose” series came
about in the late 1970s, soon after
Friedman was awarded the 1976 Nobel
Prize in economics. Friedman was
approached by a former public television manager, Bob Chitester, with the
idea of a program about his economic
and social perspectives. The result
was 10 unscripted, one-hour episodes
in which he discussed topics such as
education, protection of workers, and
inflation. For example, on the topic
of education, Friedman argued that
parents having responsibility for their
children’s education aligns with the

tradition of a free society. He argued
that elementary and secondary education in the United States should be
largely privatized and allow for the
development of a for-profit education industry to promote competition in public schools. He maintained
that providing a universal education
voucher would help bring about the
transfer of education from the government to private entities.
“Milton Friedman spent 65 years
preparing for that TV series,” Chitester
recalled. “Every step of his life he had
been preparing for that and thinking
through ideas, researching them, developing his view of the world.”
The television-book approach proved
effective: The series drew millions of
viewers and the book, co-authored
with Rose, was No. 1 on the New York
Times list of bestselling nonfiction for
six weeks.
AN ECONOMIST’S LEGACY
Friedman’s research and outreach
was consequential for the economics profession and for the country as a
whole. Some of the major policies that
he advocated were adopted more or
less in their entirety, including floating
exchange rates, the all-volunteer military, and, in some states and localities,
school vouchers.
In macroeconomic policy, the importance of monetary policy to economic
stability is widely recognized. This
understanding enabled the Fed to bring

about the Great Moderation — 40 years
of price stability in the United States
starting in the early 1980s. (The Fed
does not follow Friedman’s proposed
rule of targeting a rate of growth in the
money supply; rather, the Fed — like
many central banks — instead targets a
rate of inflation.)
Despite Friedman’s enormous influence in economics, there are areas in
which, even 18 years after his death
in 2006, the extent of his legacy is
still unclear. Among these is the
nature of corporate responsibility:
The ideas in his 1970 New York Times
essay remain controversial, and in
recent years, activist investors and
others have pushed companies to act
on environmental and social issues.
The Business Roundtable, a group
of large company CEOs, released
a statement in 2019 reversing the
group’s longtime support for principles of shareholder primacy.
On the occasion of Friedman’s
death, Fed Chair Bernanke returned
to the subject of his impact. “Among
economic scholars, Milton Friedman
had no peer,” Bernanke wrote. “The
direct and indirect influences of his
thinking on contemporary monetary
economics would be difficult to overstate. Just as important, in his humane
and engaging way, Milton conveyed
to millions an understanding of the
economic benefits of free, competitive
markets, as well as the close connection
that economic freedoms bear to other
types of liberty.” EF

READINGS
Bernanke, Ben S. “On Milton Friedman’s Ninetieth Birthday.”
Speech at the Conference to Honor Milton Friedman, University
of Chicago, Nov. 8, 2002.

Humphrey, Thomas. “The Quantity Theory of Money: Its
Historical Evolution and Role in Policy Debates.” Richmond Fed
Economic Review, May/June 1974, vol. 60, pp. 2-19.

Burns, Jennifer. Milton Friedman: The Last Conservative. New
York, NY: Farrar, Straus and Giroux, 2023.

Meltzer, Allan H. “Choosing Freely: The Friedmans’ Influence
on Economic and Social Policy.” In Wynne, Mark A., Harvey
Rosenblum, and Robert L. Formaini (eds.), The Legacy of Milton
and Rose Friedman's Free to Choose, Proceedings of a Conference
Sponsored by the Federal Reserve Bank of Dallas, October 2003.

Hetzel, Robert L. “The Contributions of Milton Friedman to
Economics.” Richmond Fed Economic Quarterly, Winter 2007,
vol. 93, no. 1, pp. 1-30.

econ focus

• third quarter • 2024 21

INTERVIEW

Raghuram Rajan
On leading a central bank, creating a digital
payment system, and India’s future in
professional services

I

22

econ focus

• third quarter • 2024

EF: When you were governor of the Reserve Bank of
India, inflation fell from a little under 10 percent in
September 2013, when you arrived, to under 4 percent in
July 2015. How did you accomplish this, and what worries
did you have along the way?
Rajan: Well, the truth is that you put in place a bunch of
measures and you hope it works. Exactly which measure
worked is hard to say. The first thing we did was that we said
we would have a glide path toward an inflation range, after
which we would think seriously about implementing inflation
targeting. We didn't want to announce inflation targeting up
front, but we wanted to make sure we could bring inflation
down to within the range we wanted to be in. And then we
could say, OK, now we will implement the targeting.
That announcement, I think, carried some weight. I
think the fact that we were serious about inflation was
further enhanced by moving from targeting the producer
price index, which reflected a lot of imported inflation
and commodity inflation. Consumer price inflation is what
people experience. So we said we were going to target
consumer prices rather than producer prices, which sounds
innocuous, but it made a huge difference because that was
what really affected people and was much higher than the
producer price, typically.
Then we did the usual central banking move of raising interest rates — all the while saying we are determined
about inflation, and as we see inflation coming down, we
will have room to cut rates. On the external side, the rupee
had been very weak; India was considered one of the fragile
five after the taper tantrum in the United States following
Chair Bernanke’s remarks in 2013.

i m age : co u rt e sy a n n e rya n

n August 2005, at the annual conference of central
bankers in Jackson Hole, Raghuram Rajan created a
stir. Rajan, then chief economist of the International
Monetary Fund, argued in a presentation that a hidden
danger of massive failures was lurking in the global
financial system. Risks had been building up, he said, a
result of the incentives facing private institutions in the
environment of that era.
Attendees were generally unmoved, if not derisive.
“The press thought I was a little bit of a crackpot,” Rajan
remembers. “There wasn’t much attention paid. It was,
‘Oh, yeah, somebody claiming the end of the world is
near.’”
Two years later, Rajan’s warning was borne out as the
global financial crisis hit and economies cratered. His prescience garnered him, among other things, an appearance
in the Oscar-winning 2010 documentary Inside Job.
Rajan later served from 2013 to 2016 as head of India’s
central bank, the Reserve Bank of India. Today, he’s a
finance professor at the University of Chicago’s Booth
School of Business. Some of his recent research has considered the implications of central banks maintaining
large asset holdings — as in the case of the Fed’s quantitative easing program — and the effects of shrinking those
holdings; other recent work of his has looked at the Indian
banking system and at unintended consequences of political pressure on monetary policymakers. He is a former
president of the American Finance Association and is a
member of the American Academy of Arts and Sciences
and the international group of economics and finance
experts known as the Group of Thirty.
He is the author or co-author of seven books. His
most recent, Breaking the Mold: India’s Untraveled Path
to Prosperity, was published in May by the Princeton
University Press.
David A. Price interviewed Rajan by phone in May.

And so, we also said we are a stable
country. We announced a scheme by
which investors could bring money into
the country in bank deposits. That was
a popular program; it raised something
like $30 billion, but also assured the
markets that we had plenty of foreign
exchange and could call on more when
we needed it. That helped stabilize the
rupee.
All in all, the package worked.
Which part specifically worked best, I
don’t know.
CENTRAL BANKS AND PAYMENT
SYSTEMS
EF: Outside of dealing with inflation
and monetary policy, you involved
the Reserve Bank of India heavily in
extending access to banking services
to individuals who lack them. Why
did you believe this was important?
And why did you believe this was
part of the role of the central bank?
Rajan: In India, the central bank has
always played a developmental role in
addition to a monetary role. So financial sector reform has often been
driven by the central bank. The RBI,
for instance, identifies priority sectors
where more lending would be useful to
reach excluded sectors of the economy.
And it mandates a certain amount of
lending to those excluded sectors.
Now, this is the historical role. And
while it is important to create the environment for easier lending, I think
you have to try and see how we can
particularly reach some of these people
and sectors that have been out of the
mainstream.
One initiative, which was driven by
the prime minister, was to get everyone
bank accounts. And given that a large
part of the Indian banking sector is
state-owned banks, it amounted to just
fiat. The prime minister said, we want
every bank to open accounts for everybody in their catchment area. And that
was a huge success in increasing the
number of bank accounts. But the next
step was to make sure that people used

their bank accounts; it’s all very easy
opening the account, but then if they
leave it dormant, you haven’t improved
banking services.
So over time, we worked on improving digital payments so that people
could use their bank accounts at a
distance. That was the beginning of
what is called the UPI, the Unified
Payments Interface, which allowed
any financial institution that was in
the network to allow their members to
make payments from any bank account
they held to any target bank account.
And that bank account to bank account
transfer was easily accomplished — so
much so that in February this year,
there were 12 billion transactions.
Digital payments also helped with
credit. Once people used their bank
accounts, once businesspeople had
records of transactions going in and
out, even the street vendor could basically show a financial institution,
“Here are the flows into my accounts
from the payments that are coming in,
so you can see how much in revenues I
make. So I am more creditworthy than
you think.” I think low-cost digital
payments coupled with near-universal
bank accounts helped propagate inclusion quite a bit.
EF: Were privacy concerns a stumbling block at the time?
Rajan: No. There were all the usual
concerns with any digital transaction
— data protection, privacy, security,
protection against cybercrime, all that
stuff. Fortunately, we had an organization, which was set up by the RBI
way back and now was owned by the
banks, called the National Payments
Corporation of India, which was tasked
with bringing new technology to
payments and implementing that. They
were very efficient.
The role of the RBI was really to
ensure that we were satisfied with the
checks and balances in their process.
Perhaps the most important thing we did
was to allow nonbanks into the process.
The banks were very reluctant to allow

the nonbanks in. We felt that the banks,
which controlled this payment interface,
would protect their own franchise and
not let the service expand. So when we
allowed the nonbanks in, that made a
huge difference.
When I last checked, three nonbanks
— Google Pay, Walmart’s Indian sub
called PhonePe, and an Indian entity
called Paytm — accounted for 95
percent of the market in UPI transactions. Almost surely, the banks would
not have been as competitive or innovative and UPI may not have taken off
if we had left it to them. There are now
worries that these new guys dominate
too much. But that’s another story.
EF: Over the course of your three
years as governor of the RBI, what
did you learn that you wish you'd
known before?
Rajan: It wasn’t so much learning big
things as trying to figure out why there
was a certain way of thinking in the
Reserve Bank. I was an outsider, and
I obviously brought a lot of academic
thinking, but I also brought impatience
with bureaucracy into the organization.
And given that it’s a hierarchical organization, like most central banks, it would
have been easy to say “jump” and people
would have done that — maybe grumbling, but they would have jumped.
The more important task was to find
out on every issue what was the thinking, what was the experience, why were
they reluctant to do A but happy to do
B. For every issue we needed to deal
with, I set up a group that was tasked
with figuring it out. The group typically had a lot of insiders. The agenda
was typically something that required
change. They all knew I wanted change
and reform. But they also knew that
I would listen to sound arguments
explaining why it was hard or it was not
advisable to move in that direction.
And they could craft the way they
wanted to change. I think that created
a lot of ownership, and it moved the
reform in interesting ways that I could
never have thought of on my own. If
econ focus

• third quarter • 2024 23

INT E RVIE W
my original thoughts had prevailed, it
would possibly have been a disaster. So
the whole issue was to learn but learn
in a way that they knew the ultimate
goal was change, because we needed to
keep reforming to improve the system.
They also knew that we would, where
possible, experiment. And if it didn’t
work, we would keep changing until it
worked. Give it some time, understand
why it’s not working, make the changes
necessary to make it work better. So we
did accomplish a lot. But most important is that there was local ownership.
And that continued when I left.
GROWTH PATHS FOR INDIA
EF: In your new book, Breaking
the Mold: India’s Untraveled Path
to Prosperity, you argue that the
conventional wisdom about developing countries — to start at the low
end with exports of commodities
and low-skilled manufacturing and
work their way up — isn’t the best
path for India. Instead, you argue
that India should seek to leapfrog
over that process as much as possible by increasing its targeting of
high-skilled services such as financial
analysis, consulting, and software.
What are the benefits and risks to
such a strategy?
Rajan: The underlying idea is that
India’s biggest asset is its human capital. And regardless of how India grows,
it needs stronger, more capable, better
human capital, especially in a world
where AI and so on are coming in in
a big way. Then the question is, where
can this human capital be used? The old
tradition was export-led manufacturing
growth: Start with low-skilled manufacturing and move up that ladder.
The problem with that, however,
is that the rents from manufacturing, especially the low-skilled assembly work, have been competed away.
Today, when you enter that area, you’re
not competing with well-paid Western
workers. You’re competing with
Chinese workers who are bolstered
24

econ focus

• third quarter • 2024

Raghuram Rajan
■ present position

Katherine Dusak Miller Distinguished Service
Professor of Finance, University of Chicago
Booth School of Business
■ selected past positions

Governor, Reserve Bank of India (2013–
2016); Vice Chairman of the Board, Bank
for International Settlements (2015–2016);
Economic Counselor and Director of
Research, International Monetary Fund
(2003–2006)
■ education

Ph.D. (1991), Massachusetts Institute of
Technology; MBA (1987), Indian Institute of
Management (Ahmedabad); B. Tech. (1985),
Indian Institute of Technology (Delhi)
by a superb infrastructure as well as
good machinery. Or you’re competing against Bangladeshi or Vietnamese
workers who are not very different
from you.
So competition is fierce at that level.
The virtuous circle by which you once
made profits from your cheap labor,
reinvested it in improving human capital, with the profitable firms paying
taxes and the government using revenues to improve infrastructure, that
part is much harder now. Now if you
want to leapfrog, you can leapfrog to
high-skilled manufacturing or you can
leapfrog to high-skilled services as the
leading sector of the economy. The
problem with high-skilled manufacturing — chip manufacturing, for example
— is that it’s hugely capital intensive.
On the other hand, high-skilled
services is not that capital-intensive. It’s human capital intensive. And
India has a lot of that human capital. India exports about 5 percent of
global services, less than 2 percent of
global manufacturing. The graduates
of India’s top universities today can
walk into a McKinsey or a Bain and do
consulting not just in India but across
the world. And you can see a whole
horde of multinational firms starting what are called global capability

centers. JPMorgan Chase hires lawyers
in India to support its operations
globally.
What this is saying is services have
become much more exportable. Now,
that doesn’t mean that’s only what you
do. To the extent that lower-skilled
manufacturing is possible, you do that.
But what I’m saying is that the manufacturing-led exports path is no longer
the only path to development. You can
have a high-skilled-services led export
path; you can have a mix. The important thing to do is to improve your
human capital, make it easier to do
business, encourage entrepreneurship
and innovation, improve your universities, improve your colleges. All this will
get you on a path for growth, which
may not be the ones that China or
Vietnam choose.
EF: Should Indian policymakers in
such a case be concerned that AI
might limit the growth of the country's exports of high-skilled services?
Rajan: It will limit the exports of
anything, right? Good AI in manufacturing could create much better robots.
That’s going to displace manufacturing workers also. So I don’t think AI is
a reason to be worried about services
in particular. I think what’s important
is how we use AI. We may enhance the
quality of services. Today, AI can help
increase the productivity of software
developers significantly. And so we
absolutely must do that.
It certainly will create some job
displacement. Maybe five, 10 years
from now, you will have AI displacing
the consumer services person you get
at the end of the line. We’re not there
yet, but we will get there. But it will
also create new jobs. AI needs prompting, for example, and people are learning how to prompt it to get the right
answers rather than hallucinations.
There’s a lot of work to do. I think
that almost surely the more skilled,
educated, creative your workforce is,
the more it can ride on AI rather than
be swamped by it.

THE ROLE OF THE INDIAN
INSTITUTES OF TECHNOLOGY
EF: The Indian diaspora has been
important to America and the
American economy. Of note, Indian
immigrants have assumed the chief
executive role at a number of major
U.S. companies, including Google,
Microsoft, and Starbucks. In an
interview in April, the U.S. ambassador to India, Eric Garcetti, highlighted this change in corporate
America. What’s your assessment?
Rajan: I don’t think, if you look at the
census numbers, that you would find
a disproportionate number of Indians
in top jobs relative to, for example, the number of highly educated
Indians there are in the United States.
Additionally, you are getting a selected
sample of Indians into the United
States; it’s a long way to migrate. Many
of them come as students in high-quality universities. Sundar Pichai [Google
and Alphabet CEO] is a graduate of
Stanford; Satya Nadella [Microsoft CEO]
is a graduate of Chicago Booth. They
come from excellent universities. My
sense is if you correct for all that, it’s
reasonable. I think that you are getting
the cream of the crop from India.
That said, if you pushed me into a
corner and insisted that Indians are
disproportionately represented as
CEOs, the only thing I can think of
is that India’s culture is a little more
oriented toward trying to reduce
conflict. Typically, Indians are less in
your face and maybe this is a better
disposition to deal with highly talented
individuals who we see in many of
these high-tech companies.
But that’s a hypothesis. I don't
have evidence for it, and it may be
totally wrong. I would first want to be
convinced that it is true that they’re
overrepresented. Clearly, one in six
CEOs is not Indian, which is what would
have to be true if you had Indian CEOs
based on their share of global population. Maybe relative to the population in the United States, we have more

CEOs coming from Indian origin than
from other origins. But we also need to
correct for how many are in the tech
industry, how many are highly educated,
all that. So it’s a very tentative answer.
EF: Speaking of education, you
received your undergraduate degree
in electrical engineering from one of
the Indian Institutes of Technology,
IIT Delhi. Do you think the IIT
system has played a significant role
in India's economic story?
Rajan: I think so. It certainly has
been world class in both the students
it admits, the competition it generates amongst them to learn, as well as
the quality of the faculty that you get
there. Of course, as India has tried to
expand the IIT system to create many
more IITs, it’s run into shortages of
faculty. But by and large, I think it was
an idea that came well before its time,
when India didn’t have the ability to
employ all the fine graduates that came
out from the IITs, and so many of them
ended up abroad.
I think they played an enormous
role outside the country. And then you
have the fact that a lot of faculty in
the United States came from the IITs.
Sergey Brin and Larry Page’s mentor
at Stanford, Rajeev Motwani, was an
Indian from the IITs.
This is a diaspora that has done well
and also spread the image that Indians
are capable, which is very important as
India expands in services, for example.
EF: Are IITs doing something different from, let’s say, American universities or universities someplace else?
Rajan: No, I would say what they
do get is the cream of a very select
crop. We've got the selectivity in IITs
higher than, for example, the selectivity at Harvard and Yale because
so many Indians apply to them and
many go through years of coaching
classes to write the entrance exam.
You get a very qualified and capable
intake of students. And then putting

them together and getting them to
learn from each other, getting them to
compete against each other, does some
of the magic; of course, the faculty
does the rest. But I don’t think it’s
unique. And I would say some of the
IITs would kill for the resources that
many U.S. universities have.
EF: Do you think the single exam
system is part of what’s helped the
IITs?
Rajan: It has in ensuring the ensuring
a clean admission process. One could
debate whether these single exams tend
to focus students for too long a time and
overly narrowly on the issue of learning
for the exam. And some of them are so
drained out, I understand, after taking
the exam, that once they get in, they are
unable to fully participate in the learning that takes place in the institution.
The exam was tough when I took it; it's
an order of magnitude tougher today
when so many kids want desperately to
get in because the IITs are still affordable. I think the process may have gone
past the optimum level of learning and
intensity for the exam.
SHRINKING THE FED’S BALANCE
SHEET
EF: Earlier, you noted the effects of
the so-called taper tantrum in the
United States in 2013, when financial market participants believed that
the Fed was about to curtail quantitative easing. What lessons have you
drawn from India’s experience in
that episode?
Rajan: I think the important point
was that stuff can happen at any time,
and it is best to be prepared for it. I
remember, for example, that before
the big taper tantrum in February of
that year, we were trying to persuade
the G20 that when monetary policy
turned, it could be potentially dangerous for the emerging markets in developing countries. The pushback we got
was, oh, it’s not going to turn, we’re
econ focus

• third quarter • 2024 25

INT E RVIE W
low for a long while, don't worry. Then
the Fed’s announcement of tapering
led to a sharp outflow of foreign capital from India and other markets, as
well as economic volatility in those
markets. But more important was India
was running a large fiscal deficit. The
current account deficit was also large
and inflation was high. That was a bad
set of macro indicators to have when
the market suddenly turned on you. I
think the lesson from that was be careful about departing too much from
reasonable macro indicators.
EF: What are you working on now?
Rajan: Viral Acharya, Sascha Steffen,
and Rahul Chauhan, a student here,
and I have been working on central
bank balance sheets, and what expanding and contracting them does. Is it
a fully benign process? Or when you
expand, do you get the system overly
dependent on central bank liquidity because you’ve created many more
reserves and then you find it hard to
withdraw? That creates possibly illiquidity, even though the central bank’s
balance sheet is much bigger than
when it initially started. So the inadvertent consequences of central bank
balance sheet expansion and contraction is what we’ve been looking at.
I’ve also been working on, of all
things, the 1950s drought in Texas and
how access to finance allowed communities to adapt to it by doing more irrigation and so on. It seems obvious
that access to finance should help, but

finance is also very local. Finance available in place A tended to help place A
and petered out over a distance. And so
it’s very important to have good local
financial institutions when you’re hit
by that type of event. At least historically that seems to be true. It probably applies to many emerging markets
today. But the bottom line is that
finance can help adaptation.
EF: With regard to your work on the
central bank balance sheets, the Fed
right now is in the process of gradually reducing the size of its balance
sheet. What do you think is important for the Fed to have in mind as
this process is underway?
Rajan: I think the last statement of the
FOMC to some extent mirrored our
concerns. What we’re saying is, look,
it’s not going to be easy to shrink your
balance sheet, even though it is much
bigger than when you started. And
that’s because the system has come to
depend on it.
What we’ve seen with the Fed
shrinking its balance sheet is that
initially what shrank were the reverse
repos that the Fed did with money
market funds. Those, we think, are
relatively benign. But once you start
shrinking the reserves held by the
banks themselves, it becomes a tougher
process. And so you want to proceed
slowly. Yes, you want to do it. I absolutely am for shrinking the central
bank balance sheet. But you want to do
it carefully, giving the system enough

Enjoying Econ Focus?
Subscribe now to get every
issue delivered right to your door.
Visit https://www.richmondfed.org/publications/print_subscription.

26

econ focus

• third quarter • 2024

time to react because too abrupt a shift
in the reserves outstanding can create
significant liquidity problems. At least,
that’s what the past tells us.
So I am happy that they’ve decided
to slow down the pace of shrinkage.
That means they will have more time
to observe what is going on and react
accordingly. Are the usual measures of
illiquidity starting to move up? Do you
see potential concerns about liquidity not reaching the right places, some
spreads moving up, some interest
rates moving up when they normally
shouldn’t? All those are signs that
things aren’t going well. So I think
close monitoring is warranted, and I’m
glad that the Fed is doing that.
EF: It’s been reported that you’re a
fan of J.R.R. Tolkien. Is that true,
and if it is, where did your affinity
for him come from?
Rajan: It is true. I’ve always enjoyed
deep fantasy of the kind that Tolkien
writes. I chanced upon his books in my
late teens. I just saw them somewhere
and started reading them and was fascinated. And then when my daughter was
growing up, I read the books again to
her, and she enjoyed it. And then the
movies came along, and she’s watched
them a zillion times. So the whole package is fascinating. And of course, I also
read Harry Potter to her. It was a nice
excuse to be able to read to your children because then you can relish the
books without somehow feeling that
you’re not doing the adult thing. EF

DISTRICT DIGEST
b y s t e p h a n i e n o r r i s , l a u r a d aw s o n u l l r i c h , a n d s o n ya r av i n d r a n at h w a d d e l l

Preparing to Work: Changing Demand for
Postsecondary Education?

S

tarting around a decade before the
COVID-19 pandemic, and more
acutely during the pandemic itself,
postsecondary enrollment declined
— most notably among community
colleges, both in the Fifth District and
in the United States as a whole. Yet the
2023-2024 academic year saw a shift
in that trend as enrollment grew once
again. What’s going on?
The unsatisfying answer is that
there are conflicting forces at work,
and it’s hard to tell which will prevail.
But one thing is clear: Combined with
the anticipated decline in the collegeage population, high costs of four-year
degrees, and changing demands among
employers, parents, and students,
higher education seems to be at a
crossroads. Now might be the time to
rethink the human capital needs of
America’s future workforce and the
programming required to meet them.
ENROLLMENT TRENDS
It is a long-held truth among those who
study education and the workforce that
completing a postsecondary degree will
improve an individual’s labor market
outcomes. On average, in the labor
market, a person is better off finishing high school than not finishing high
school, is better off with an associate
degree than a high school diploma, is
better off with a bachelor’s degree than
without, and is better off still with an
advanced degree. This pattern shows up
not only in earnings, but also in everything from likelihood of employment to
resilience in an economic downturn. In
June, for example, the unemployment
rate among those who only finished
high school was 4.2 percent — almost
twice as high as the rate for those
who held at least a bachelor’s degree.
At the height of the Great Recession

(December 2009), high school graduates had an unemployment rate of 10.6
percent compared to 4.9 percent for
those with a bachelor’s degree.
Of course, the “wage premium” for
higher education represents an average;
it does not tell us the return an individual will receive from attending school,
which depends not only on the individual’s interests and abilities, but also
on the type of degree, the major, and
the institution. Nonetheless, given the
overall relationship between education
and income, it is not surprising that for
decades we saw enrollment at four-year
institutions increase. We mainly focus
on public four-year, private nonprofit
four-year, and public two-year institutions, which consistently account
for most postsecondary enrollment. In
the spring of 2024, for example, those
three types of institutions accounted
for almost 90 percent of total enrollment in higher education. (There are
technical differences between the categories of public two-year institutions
and community colleges, but the vast
majority of public two-year institutions
are community colleges, and we use
the terms interchangeably here.)
Starting in about 2010, total enrollment in postsecondary pursuits started
to fall. Part of the reason was demographic: For 18-to-21-year-olds, the estimated population fell by an average of
0.6 percent per year from 2010 to 2019.
But the share of high school graduates
who chose postsecondary education also
fell. In 2022, 62 percent of high school
graduates enrolled the following fall in
a two- or a four-year college — this was
8.1 percentage points lower than in 2010
and 4.2 percentage points lower than
in 2019. Part of this might be due to the
flattening in the college wage premium
even as rising costs for higher education have driven more students into debt.

In other words, the enrollment decline,
in addition to being demographic, could
be related to decreased affordability, a
decline in the perceived value of education, or a perception of increased volatility in its value. Importantly, much
of the decline in enrollment came
from a decrease in community college
enrollment.
The emergence of the COVID-19
pandemic in 2020 introduced new challenges for enrollment, exacerbating the
already complicated enrollment environment. For example, the pandemic
necessitated a move to online classes,
created skepticism about the value of
an online degree or any postsecondary degree, and enhanced challenges
finding family care for parents seeking
to go to school. COVID-19-era shifts
were especially disruptive for disciplines that cannot be taught online,
such as welding or dental hygiene. In
many cases, these classes continued in
person, but enrollment was severely
limited to ensure social distancing. In
fall 2020 and again in fall 2021, total
enrollment fell by more than 2 percent.
The cumulative decline between fall
2019 and fall 2021 was 4.7 percent.
Again, community college enrollment struggled the most: Two-year
public institution enrollment declined
a staggering 14.4 percent, while fouryear public enrollment declined only
0.4 percent.
Recently, we have started to see
total enrollment come back (though it
remains below pre-COVID-19 levels).
However, we might be witnessing a
shift in demand for postsecondary
education.
A SHIFT IN ENROLLMENT
The 2023-2024 academic year brought
with it much-anticipated and hoped-for
econ focus

• third quarter • 2024 27

DIS T R IC T DIG E S T

Change in Enrollment by Institution Type: U.S. and Fifth District

28

econ focus

• third quarter • 2024

Year-over-year % change
6%
4%
2%

PERCENT

0%
-2%
-4%
-6%
-8%
-10%
-12%

2019-2020

2020-2021

2021-2022

2022-2023

U.S. Public 4-year

Fifth District Public 4-year

U.S. Private nonprofit 4-year

Fifth District Private nonprofit 4-year

U.S. Public 2-year

Fifth District Public 2-year

2023-2024

Enrollment Change Spring 2023 — Spring 2024
15%
10%
5%
PERCENT

increases in enrollment across all
sectors of higher education: Following
trends from the fall of 2023, the
National Student Clearinghouse (NSC)
data on enrollment in the spring of
2024 indicates that enrollment across
institutions was up. Nationally, undergraduate enrollment grew 2.5 percent
in the spring of 2024 for the second
consecutive semester following the
pandemic declines.
Enrollment growth was strongest
at community colleges, both in the
United States and in the Fifth District.
(See top chart.) In fact, bucking the
pre-pandemic and pandemic trends,
community college enrollment was the
only type of enrollment that was up in
every Fifth District state from spring
2023 to spring 2024. (See bottom
chart.)
In spite of this growth, community
colleges are still further behind their
2019 enrollment levels than four-year
institutions. In the Fifth District, for
example, enrollment at community
colleges was still about 20,000 students
(3.7 percent) below its 2019 levels in
spring 2024, while four-year public
institutions were less than 1 percent
below and four-year private nonprofits
were slightly above their 2019 levels.
Growth trends have shifted, however,
and as technology and the demands of
employers continue to develop, these
shifts will matter not only for employers and workers, but also for the institutions that may be at risk and the
students, staff, and communities that
rely on those institutions.
Looking ahead, the demographic
shifts that began over a decade ago
are about to become more severe. One
study estimates that the number of
high school graduates will peak at 3.9
million in 2025 and after that, we will
see about a 10 percent decline, such
that the class of 2037 will be about
the same size as the class of 2014.
This “2025 cliff” is a result of declining fertility rates, which became
more severe at the beginning of the
Great Recession in 2007. Colleges and
universities will be competing for

0%
-5%
-10%
-15%

Maryland
Public 4-year

North Carolina

South Carolina

Private nonprofit 4-year

Virginia

West Virginia

Public 2-year

SOURCE: National Student Clearinghouse Current Term Enrollment Estimates (spring 2024); authors' calculations.
NOTE: Includes graduate and undergraduate students. Based on spring enrollment.

an ever-smaller number of freshman
students.
Changing demand might also shift
enrollment for different types of postsecondary pursuits within institutions.
In the latest data from the NSC, not
only did we see increases in community college enrollment that outpaced
four-year institutions, we also saw
a change in the type of community
colleges that students are choosing. In
the spring of 2024, community colleges

with a larger percentage of students
enrolled in vocational programs
increased total enrollment by almost
18 percent — with enrollment that now
exceeds pre-pandemic enrollment by
4.6 percent. (See chart on next page.)
On the other hand, community colleges
that have more students enrolled in
programs designed to transfer to fouryear colleges grew enrollment in 2024,
but enrollment levels remain well below
2019 levels. Undergraduate certificate

Change in Community College Enrollment by Institutional Program Focus

HOW DO STUDENTS DECIDE?
In theory, students and their families must weigh the costs of pursuing
a program (such as tuition and fees,
forgone earnings, and the cost of child
care during class times) against the
benefits (such as the wage premium
upon graduation, a fulfilling career
in their preferred location, and prestige). Quantifying the costs is relatively
straightforward, but the benefits are
harder to project. To find out which
skills will be in high demand in their
areas, students may rely on labor market
projections for their area or simply look
to the largest local employers.
One key piece to this decision should
be the likelihood of completing the
chosen educational pursuit, be it a
degree, a certificate, or acquisition of a
certain skill. There is ample evidence
that whether you are seeking a bachelor’s degree, associate degree, or a
certificate, you are better off completing the degree. Thus, in addition to
evaluating the cost and benefit of the
educational pursuit itself, it is critical to weigh the risk of not completing. There are many factors that affect
a student’s propensity to complete an
academic program, but when evaluating their choices, many students
rely on broad metrics such as institution-specific graduation rates.
As we have written about previously,
finding appropriate metrics can be challenging for those pursuing a path at
or through a community college. For
many years, the most used definition of

Year-over-year % change
15%
10%
5%
PERCENT

programs at both community colleges
and four-year institutions are experiencing the largest increases in enrollment, and certificate enrollment has
now more than recovered from COVID19-era declines. Associate degree enrollment is growing faster than bachelor’s
degree enrollment, but enrollment for
both types of degrees lag pre-COVID-19
levels. Across higher education, there
is evidence that since the pandemic,
student demand has shifted toward
shorter-term academic programs.

0%

2019-2020

2020-2021

2021-2022
2022-2023

2023-2024

Virginia

West Virginia

-5%
-10%
-15%

Maryland

High Transfer
Public 4-year

North Carolina

South Carolina

Mixed Transfer
High Vocational
Public 2-year
Private nonprofit 4-year

SOURCE: National Student Clearinghouse Current Term Enrollment Estimates (spring 2024); authors' calculations.
NOTE: Includes graduate and undergraduate students. Based on spring enrollment.		

postsecondary success, the traditional
Integrated Post-Secondary Educational
Data System (IPEDS) graduation rate,
has best aligned with the goals of
four-year institutions — that is, defining success as degree completion by
a first-time, full-time, degree-seeking
student. We typically measure success
for community colleges and four-year
students and institutions with identical
metrics. Yet before students can decide
on their postsecondary path (and before
policymakers can make decisions about
what programs to invest in), we need to
define success for community colleges
in a way that reflects their objectives
and the populations that they serve.
COMMUNITY COLLEGES: SORRY,
WRONG NUMBERS
Community colleges play a unique role
in the U.S. higher education system.
By definition, these institutions tend
to serve a specific geography, allowing them to tailor their program offerings and support services to the needs
of their local community. Through
deep relationships with local employers and by building curricula around the
skills needed for work in local industries, community colleges often provide
a direct pipeline to the local workforce.
Community colleges also serve students

from a broad array of socioeconomic
and demographic backgrounds. They
provide students who plan to attain a
bachelor’s degree with a low-cost alternative for the first two years of college.
Nontraditional students who want to
reskill or upskill can attend community
college for short-term training opportunities. They provide opportunities for
part-time students who need to work or
care for family members while seeking
a degree. They also provide dual enrollment opportunities for high school
students. In the most recent NSC data,
the number of dual enrolled students
increased for the third consecutive year,
comprising almost 30 percent of the
undergraduate enrollment increases.
As postsecondary demand has
shifted toward community colleges, it
has become more important than ever
to understand how they are serving
their students and to define success in
a way that accurately represents their
programming. In response to these
issues, the Richmond Fed has been
engaged in an effort to rethink the
measurement of community college
success and to collect data on a more
holistic group of community college
students, including those enrolled in
non-credit programs. The Richmond
Fed’s Survey of Community College
Outcomes includes a new approach to
econ focus

• third quarter • 2024 29

DIS T R IC T DIG E S T

Comparing the IPEDS Graduation Rate and the Richmond Fed Success Rate,
By Community College
15%
90%
80%
10%
70%
PERCENT

5%
60%
50%
0%
40%
-5%
30%
20%
-10%
10%

Maryland

North Carolina

South Carolina

Virginia

West Virginia

School 1
School 2
School 3
School 4
School 5
School 6
School 7
School 8
School 9
School 10
School 11
School 12
School 13
School 14
School 15
School 16
School 17
School 18
School 19
School 20
School 21
School 22
School 23

-15%
0%

Public 4-year

Private nonprofit 4-year

IPEDS Graduation Rate

Public 2-year

Richmond Fed Success Rate

SOURCES: Federal Reserve Bank of Richmond; National Center for Education Statistics (NCES) Integrated Postseconday Data System.
NOTE: Institution names are omitted to maintain anonymity.

measuring success, which broadens
the definition of community college
student success to include not only
degree attainment, but also attainment of shorter-term credentials,
such as certificates or industry licensures, successful transfer to a four-year
institution, or persistence in enrollment beyond four years. For example,
in Virginia, although the traditional
IPEDS graduation rate does a good
job at approximating success for some
schools, success can look very different in other schools when we take
into account the full array of community college programming and define
success accordingly. (See chart.)
We also collect data on students
enrolled in non-credit programs so
that we can observe the full range
of student enrollment at community
colleges. All the prior enrollment data
presented in this article reflects enrollment only in credit-bearing academic
programs across institutional sectors.
However, a large and growing percentage of community college students
are enrolled in non-credit, workforce
programs. These programs range from
commercial driver’s license, or CDL
programs, to phlebotomy to welding. (See “Non-Credit Workforce
30

econ focus

• third quarter • 2024

Programs at Community Colleges,”
Regional Matters, Feb. 22, 2024.)
Outreach to community colleges indicates that not only is enrollment shifting from degrees to these shorter-term
programs, it is also shifting from credit
programs to non-credit programs. We
can’t observe those students in the
data from NSC or other national data
providers, though, complicating the
enrollment story.
QUANTIFYING THE BENEFITS
Defining success and understanding
the likelihood of completion is a first
step, but as individuals negotiate labor
market changes and seek to maximize their investment in postsecondary
training, there is a renewed sense of
urgency for quality information on the
payoff to different educational pathways. Why is it so hard to link specific
higher education choices to labor
market outcomes?
These data are inherently difficult to
produce because it requires following
students’ post-degree skills attainment
and because it is difficult to control for
the range of factors outside of their
educational attainment that can affect
their earnings. While we know that

college graduates have lower unemployment rates than noncompleters,
this varies by field of study and degree
attained. The New York Fed’s labor
market data on recent college graduates
show that while the unemployment
rate in 2022 for those with a bachelor’s degree or higher was 2.2 percent,
this ranged from 0.2 percent for those
with an industrial engineering degree
to 8 percent for those with an art
history degree. Not surprisingly, wages
differ as well. While those with industrial engineering degrees between the
ages of 22 and 27 had a median wage
of $71,000, those with an art history
degree had a median wage of $41,000
at the same age.
To improve data in this area, the
Census Bureau has been working on
a project known as Post-Secondary
Employment Outcomes (PSEO). The
PSEO data include earnings and
employment outcomes for community college and four-year graduates
by degree level, major, and institution
attended. Not every state has chosen
to participate, but two Fifth District
states — South Carolina and Virginia
— have participated. The data indicate
that there are some industries, such
as health professions, where wages
increase dramatically with attainment
of an associate degree, but further
degree attainment doesn’t result in
notably higher wages one year after
award completion. In others, such
as engineering and business, attaining a bachelor’s or master’s degree in
the field results in significantly higher
wages, on average.
Measuring the value of non-credit
certificates, licenses, or third-party
credentials is even more challenging. Students who attend short-term
training programs are often excluded
from enrollment and graduation rates
and data on third-party credential
attainment is very difficult (or even
impossible) to attain. Until there are
standardized data on enrollment in
non-credit workforce programs and
the tangible outcomes that students
attain, even the most robust datasets

linking educational outcomes to wages
will be limited in their ability to
reflect community college outcomes.
There are, of course, benefits to
higher education that are not directly
reflected in wage data, and low wages
don’t always indicate low demand
(or low value to society). Child care
workers, for example, are in demand
nearly everywhere, but the wages
in the child care industry remain
very low. Wages also reflect individual characteristics and preferences
that are independent or only tangentially related to their field of study and
occupation. Additionally, many people
work in occupations that are unrelated to their degree or certificate.
Still, knowing the earnings potential
of different educational and career
paths is important for students seeking to make sound economic decisions
around how they spend their postsecondary training.
FOR SOME INSTITUTIONS, A SQUEEZE
As the pool of high school graduates
shrinks and a smaller number of new
first-time students enroll in college
each year, some schools will feel the
effects more sharply than others. In
about the mid-2010s, market forces
were already putting small colleges like
Sweet Briar College in Lynchburg, Va.,
at risk of closing — though Sweet Briar
recovered thanks to the rallying of
alumnae. (See “Too Small to Succeed?”
Econ Focus, First Quarter 2017.) More
recently, schools as different as the
University of Lynchburg and West
Virginia University have announced
major changes to offerings.
In general, flagship universities have
maintained increased enrollment, but
many regional public colleges and
universities have experienced declines.
For example, in South Carolina, undergraduate enrollment at Clemson and
the University of South Carolina
(USC)-Columbia grew by 32 and 13.8
percent, respectively, between 2013
and 2022, while undergraduate enrollment at some smaller regional public

universities, such as USC-Upstate and
Winthrop University, saw declines of
more than 20 percent.
Similarly, elite private colleges and
universities have welcomed record-setting classes, while other private schools
have seen persistently declining enrollment. For example, Washington and
Lee, in Lexington, Va., an academically elite private university, saw
enrollment grow from 1,845 full-time
students in fall 2019 to 1,859 full-time
students in fall 2022, maintaining their
preferred institution size. Over the
same period, Marymount University,
a similar-sized private nonprofit institution in Arlington, Va., saw full-time
undergraduate enrollment fall from
1,951 to 1,644, a 15.7 percent decline.
These institutions also depend on
tuition and enrollment differently. As of
June 30, 2022, Marymount University’s
endowment had a balance of $49.2
million, approximately $30,000 per fulltime enrolled undergraduate. As of the
same date, Washington and Lee had an
endowment balance of about $2 billion,
approximately $1.1 million per full-time
enrolled undergraduate. Schools like
Marymount are often called “tuition-dependent” because they rely on tuition
revenue to meet annual expenses and do
not have significant endowment income
to help weather periods of enrollment
declines or cost increases.
Community colleges face the same
enrollment environment as four-year
institutions without the same level of
state funding or large endowments.
However, their unique position within
higher education and workforce development that has disadvantaged them
in outcome and workforce metrics
may serve them well, as labor force
demands and student preferences shift.
For one, community colleges can be
a low-cost option in an environment
where students and parents are increasingly questioning the value of a higher
degree. In addition, community colleges
are more adept at shifting to meet the
needs of their local workforces.
Of course, as demand for shorter-term degrees rises, there is room for

community colleges to better align their
educational services with the needs of
the local workforce. Recent research
from the Georgetown University’s
Center on Education and the Workforce
suggests misalignment in many communities between associate degrees and
certificates earned and the skills needed
for occupations that are increasingly
in demand. Community colleges with
direct lines of communication with
employers and partnerships can pivot
to train in-demand workers through
non-credit short-term credential
programs. Of course, if policymakers
wish for community colleges to fill this
role in the workforce ecosystem, clarity around outcomes, aligned incentives,
and funding to provide these services
will be critical.
CONCLUSION
There will be changes to demand
for higher education that come from
long-term trends such as demographic shifts or technology like artificial intelligence that changes labor
market demands. There will also be
short-term changes that might have
long-term implications. For example,
the increasing enrollment trends at
community colleges relative to fouryear institutions may well be boosted
by this year’s FAFSA debacle in which
the Department of Education repeatedly delayed release and processing
of the new federal financial aid application. (See “June Update: The 2024
FAFSA Crisis,” Community College
Insights, June 21, 2024.) But most
critically, for students to make the
right decisions for themselves, they
need good information about how
different institutions or types of postsecondary education will enhance
their longer-term prospects. To understand which programs to invest in,
our policymakers need to understand
the current and future labor market
outcomes from different postsecondary programs. No one has a crystal
ball, but better data and more research
could help. EF
econ focus

• third quarter • 2024 31

OPINION
by z h u wa n g

Artificial Intelligence: Potentials and
Prospects

W

e are at the dawn of a new technological revolution.
The recent development of artificial intelligence
(AI), especially the emergence of generative AI, has
offered a plausible future in which machines will eventually
free humans from a wide range of cognitive tasks, unleashing vast creativity and productivity gains.
Historically, AI technologies have progressed gradually,
through cycles of optimism and disappointment. In recent
years, however, the use of AI and machine learning technology has started gaining ground in various applications, such
as search engines, targeted advertising, self-driving vehicles, language translation, and image recognition.
The most impressive leap is the rise of generative AI,
marked most notably by the release in November 2022 of
ChatGPT, which can generate text, code, images, and other
data, often comparable to or surpassing human quality. The
latest models of generative AI have demonstrated the abilities of solving novel and difficult tasks that span mathematics, coding, vision, medicine, law, and other areas, and they
continue to improve at a fast pace. The use of generative AI
is on the way to transforming a large variety of industries,
including finance, software development, customer service,
health care, entertainment, sales and marketing, art, writing, fashion, and product design, and the list is growing.
While the future of AI is thrilling, there are important
questions about how to best harness the potential of AI and
prepare for the challenges and risks along the way. In that
regard, economic history and research can provide some
useful thinking.
First, it may take a long time to achieve measurable,
large-scale productivity gains from AI. History has shown
repeatedly that revolutionary technological advancements
often come with a “productivity paradox.” At the turn
of the 20th century, with the early adoption of electrical
power, engineers were envisioning profound transformations enabled by electrification, but that vision did not materialize until two decades later, when electrification finally
attained a 50 percent adoption level among U.S. households
and manufacturing plants. Similarly, the increasing adoption of computers did not result in the widely anticipated
productivity surge in the 1970s and 1980s; as Robert Solow
remarked in 1987, “You can see the computer age everywhere but in the productivity statistics.”
A fundamental reason for this delay is that it takes time and
resources to develop complementary inputs associated with
a technological breakthrough, including co-invention of new
processes, products, business models, and human capital. The
32

econ focus

• third quarter • 2024

more revolutionary the technology advance, the more complex
and costly the transition can be. This could show up as a slowdown or stagnation in productivity growth, and the benefits
would not be harvested until years or decades later. AI is likely
to be at the early stage of such an evolutionary path.
Second, technology changes can have a big impact on jobs
and income distribution. Much as automation has replaced
manual labor on the factory floor, AI can take over tasks
from knowledge workers. In an optimistic scenario, this
may enhance the productivity of knowledge workers or
even move them up to more creative and better-paid jobs.
But in a pessimistic scenario, AI may substitute knowledge workers or relegate them to less productive, lowerpaid positions. Also, depending how AI is introduced
and deployed, it could widen or shrink the digital divide
between those who are privileged and those who are not,
and this could have profound consequences on economic
and social inequality.
Third, technology is a double-edged sword that can be
misused. Nuclear technology is a familiar example that can
be used for both beneficial applications, such as nuclear
energy and nuclear medicine, and mass destruction. As the
potential of AI continues to unfold, there are also growing concerns about harmful uses of AI. For example, AI can
be abused to generate fabricated stories and fake images
to spread misinformation; AI systems trained on biased
or incomplete datasets can perpetuate societal biases and
discrimination; and large-scale adoption of defective or
malign AI algorithms can elevate systemic risks. For many,
there is also an ultimate worry that unsupervised AI
advances may create a superintelligence conflicting with
human values that could lead to catastrophic outcomes,
even possibly human extinction.
These thoughts and concerns highlight the potentials
and perils of AI. They also point to the important roles
that public policies can play in guiding AI development
and implementation. It is essential for researchers, business practitioners, policymakers, and the general public
to work together to develop effective policies and robust
regulation to coordinate and facilitate the continuing
progress and adoption of AI and address potential downside risks. AI has fantastic potential and needs to be developed and used responsibly for the benefit of all. EF
Zhu Wang is vice president for research in financial and
payments systems in the Research Department of the
Richmond Fed.

Federal Reserve Bank
of Richmond
P.O. Box 27622
Richmond, VA 23261
Change Service Requested

To subscribe or make subscription changes, please email us at research.publications@rich.frb.org.

Calling All College Students!
Save the Date: Diverse
Economics Conference
Oct. 11, 2024
Are you an undergraduate student interested in an economics career?
Join us Oct. 11 in Richmond, or virtually, for the fifth annual DivEc Conference.
Whether you’re drawn to environmental sustainability, technological advancement,
advocating for issues related to race and gender, or exploring the endless array of
other possibilities, this conference will empower you with the knowledge, skills, and
connections to forge your unique path toward meaningful change.
Through an engaging keynote speech, interactive panel discussion,
and immersive poster session, you’ll hear from experts in the field on
how they’ve leveraged their passions into fulfilling careers.

Scan here to learn more and register to attend.
Seats fill up fast, so don’t wait too long!