The Global Economic Recovery 10 Years After The 2008 Financial Meltdown

2
CHAPTER
THE GLOBAL ECONOMIC RECOVERY 10 YEARS AFTER
THE 2008 FINANCIAL MELTDOWN
This chapter takes stock of the global economic recovery a ing panic—marked by distressed asset sales, deposit
decade after the 2008 financial crisis. Output losses after withdrawals from banks and money market funds,
the crisis appear to be persistent, irrespective of whether and the freezing of credit—triggered a collapse in
a country suffered a banking crisis in 2007–08. Sluggish cross-border trade and led to the worst global recession
investment was a key channel through which these losses in seven decades.
registered, accompanied by long-lasting capital and total Ten years later, the sequence of aftershocks and
factor productivity shortfalls relative to precrisis trends. policy responses that followed the Lehman bankruptcy
Policy choices preceding the crisis and in its immedi- has led to a world economy in which the median
ate aftermath influenced postcrisis variation in output. general government debt-GDP ratio stands at 52 per-
Underscoring the importance of macroprudential policies cent, up from 36 percent before the crisis; central bank
and effective supervision, countries with greater finan- balance sheets, particularly in advanced economies,
cial vulnerabilities in the precrisis years suffered larger are several multiples of the size they were before the
output losses after the crisis. Countries with stronger crisis; and emerging market and developing econo-
precrisis fiscal positions and those with more flexible mies now account for 60 percent of global GDP in
exchange rate regimes experienced smaller losses. Unprec- purchasing-power-parity terms (compared with 44 per-
edented and exceptional policy actions taken after the cent in the decade before the crisis), reflecting, in part,
crisis helped mitigate countries’ postcrisis output losses. a weak recovery in advanced economies.
Against this backdrop, this chapter takes stock of the
global economic recovery 10 years after the financial
Introduction meltdown of 2008 and the policy lessons that can help
Over the weekend of September 13–14, 2008, two prepare for the next downturn. Specifically, the chapter
large US financial institutions teetered close to failure addresses the following questions:
while a third urgently sought a buyer to avoid that same • Compared with precrisis trends, how did output
fate. By Sunday night that weekend, Merrill Lynch was evolve across countries in the aftermath of the crisis?
acquired by Bank of America. Insurance giant AIG still • How did the associated components—capital, labor
desperately pursued credit lines, just days away from a inputs, total factor productivity (TFP)—advance
ratings downgrade that looked likely to push it over the after the crisis? What does this decomposition show
edge. And in the early hours of Monday, September 15, about why it took a long time for output in many
2008, the investment bank Lehman Brothers filed for economies to return to its precrisis level?
bankruptcy, brought down largely by its exposure to a • Even as the world economy experienced its worst
US housing market in deep decline. slump in seven decades, postcrisis macroeconomic
The post-Lehman scramble for liquidity in global performance varied across countries. What accounts
markets heralded the most acute phase of the financial for this variation? Which policies and structural attri-
turmoil that, by then, had been brewing in the United butes helped limit the damage and facilitate recovery?
States and Europe close to 18 months.1 The ensu-
The chapter uses a sample of 180 countries—
The authors of this chapter are Wenjie Chen, Mico Mrkaic, covering advanced, emerging market, and low-income
and Malhar Nabar (lead), with contributions from Deniz Igan,
Christopher Johns, and Yuan Zeng, and supported by Luisa Calixto,
Meron Haile, and Benjamin Hilgenstock. mortgage-related hedge funds associated with Bear Sterns (June
1Identifying a precise starting point for the timeline—the “patient 2007) and BNP Paribas (August 2007); the United Kingdom’s first
zero” of the epidemic—is difficult. This chapter takes the April bank run since the 19th century, on Northern Rock (September
2007 collapse of subprime mortgage lender New Century Finan- 2007); the failure of mortgage lender Countrywide Financial (Jan-
cial as the first major distress sign following the mid-2006 turn in uary 2008); JPMorgan’s acquisition of Bear Sterns with US Federal
the US housing market. Key markers of financial stress over the Reserve support (March 2008); and the US government’s takeover of
subsequent 18 months include the suspension of redemptions from mortgage giants Fannie Mae and Freddie Mac (September 2008).
International Monetary Fund | October 2018 71

WORLD ECONOMIC OUTLOOK: Challenges to Steady Growth
developing economies—to quantify output losses, • Financial: Underscoring the importance of macro
explore the precrisis correlates of postcrisis variation prudential policies and effective supervision, the
in output performance, and examine whether actions analysis finds that countries in which financial
taken in the immediate aftermath of the crisis are vulnerabilities had accumulated to a larger degree
associated with limiting output losses over the medium in the precrisis years suffered greater output losses
term (2015–17). Previous World Economic Outlook after the crisis. In the years running up to the crisis,
(WEO) analysis (October 2009) examines output per- countries with larger excess current account deficits
formance after an earlier set of financial crises during and those with more rapid credit growth found
1970–2002. The current chapter builds on that by that constraints bound relatively more strongly
zeroing in on the aftermath of the 2008 crisis. when financial conditions tightened after the crisis.
An important consideration when comparing pre- Stricter banking regulation (proxied by an index of
and postcrisis output patterns is the extent to which restrictions on certain aspects of bank activity) in
precrisis growth was fueled by excessive credit growth the precrisis years is associated with a lower proba-
and unsustainable investment that had to be worked off. bility of a banking crisis in 2007–08.
A related issue is whether structural change unrelated • Policy constraints and frameworks: The evidence
to the crisis may have affected trend growth over time suggests that countries with stronger precrisis fiscal
in some countries (specifically, whether some coun- positions experienced smaller output losses in the
tries experienced temporarily elevated potential growth aftermath. The analysis also finds that flexible
rates before the crisis that subsequently reverted to the exchange rate regimes helped lessen GDP damages.
long-term average). As discussed in the next section, the • Postcrisis actions: Several countries took unprece-
analysis attempts to adjust precrisis trends for the influ- dented and exceptional policy actions to support
ence of factors, such as credit growth, that may affect the their economies after the 2008 financial meltdown.
path of output beyond the influence of typical demand The chapter finds that these actions (specifically,
fluctuations. Even with this correction, for some coun- quasi-fiscal measures to support the financial sector,
tries, the output deviations from precrisis trends may still including guarantees and capital injections) helped
capture the effect of slow-moving structural changes in temper postcrisis output losses.
trend growth rates over time. Nonetheless, the chapter’s
cross-country analysis—comparing countries that expe- Some of these factors appear to be particularly
rienced banking crises in 2007–08 with those that did relevant for the euro area. The 2008 financial crisis
not, as well as across income levels—can help identify exposed thin buffers in some member economies and
precrisis drivers of postcrisis output deviations. gaps in the architecture of the currency union. The
Among the main findings of the analysis are that interaction of domestic and area-level factors exacer-
output losses appear to be persistent and not restricted bated adjustment difficulties in the euro area follow-
to countries that suffered a banking crisis in 2007–08. ing the 2008 shock and gave rise to an intensifying
Sluggish investment appears to be a key channel sovereign debt crisis during 2010–12, which spurred
through which these losses registered, with associated efforts to strengthen the architecture of the currency
long-lasting capital and TFP shortfalls relative to their union (IMF 2012, 2013a; Allard and others 2013;
precrisis trends. Consistent with these TFP shortfalls, Goyal and others 2013; Berger, Dell’Ariccia, and
research and development expenditure and technol- Obstfeld 2018). In contrast to the 2009 shock, euro
ogy adoption appear to have increased more slowly area countries hit by the sovereign crisis were not in
in countries that suffered larger output losses. The a position to use expansionary fiscal policy to counter
findings are similar to those of recent papers showing the “sudden stop.” Rather, they needed to reduce
that output tends to stay below previous trends after their fiscal deficits to regain creditors’ confidence
crises and recessions (for example Cerra and Saxena and contain sovereign borrowing costs. In the event,
2008, 2017; Blanchard, Cerutti, and Summers 2015; the contractionary effect of this fiscal tightening was
and Aslam and others, forthcoming). larger than anticipated at the time (Blanchard and
The analysis finds that policy choices leading up to Leigh 2013; IMF 2013b, 2015).
the crisis and in its immediate aftermath influenced The next section quantifies the losses in output and
postcrisis variations in output performance. These can discusses the channels through which they occurred.
be grouped into three categories. The subsequent section examines the policy and
72 International Monetary Fund | October 2018

CHAPTER 2 THE GLOBAL economic RECOVERY 10 YEARS AFTER the 2008 FINANCIAL MELTDOWN
structural attributes that, in part, account for variation Figure 2.1. Correlation of GDP Deviations between Periods
in postcrisis output. The main takeaways are summa- (Percent)
rized in the conclusion.
Postcrisis performance is persistent, with a correlation coefficient between GDP
deviations for 2011–13 and 2015–17 of about 0.90.
Persistent Post–Global Financial Crisis
Deviations in Output 40
Following the global financial meltdown in late

2008, 91 economies, representing two-thirds of global 20
GDP in purchasing-power-parity terms, experienced
a decline in output in 2009. By way of comparison, 0
during the 1982 global recession, 48 economies,
accounting for 46 percent of world GDP, registered
2015–17
–20
output declines compared with the previous year.
To get a sense of the long-lasting changes in output
after the 2008 crisis, this chapter measures postcrisis –40
deviations of output from the level that would have
prevailed had output followed its pre‑2009 trend –60
growth rate (Ball 2014). Considering that generally
accommodative financial conditions likely contrib-
–80
uted to unsustainable growth in many countries prior –30 –20 –10 0 10 20 30
to 2008, it is important to adjust for these influ- 2011–13
ences when estimating an underlying trend path for

Source: IMF staff calculations.
output as the benchmark for comparison (Online Note: GDP deviations are average percent deviations from precrisis trend.
Annex 2.2.B).2,3 Nevertheless, despite this adjustment,
in some cases, the measured output deviations may
include country-specific changes in trend growth rates
The post-2008 output deviations exhibit strong
that are unrelated to the crisis. Consider the world’s two
persistence over time (Figure 2.1).5 A second note-
largest economies, for example. In the United States, a
worthy aspect is that economies with larger output
slowdown in total productivity growth that predates the
and employment losses in the initial aftermath of the
2008 crisis has contributed to lower potential growth
crisis registered greater increases in income inequality
over time (Fernald 2015; Adler and others 2017).
compared with their precrisis average (Figure 2.2).6
China’s economy has experienced major structural shifts
These developments help shed light on the lingering
that span the 2008 crisis and an associated transition
sense of subpar economic performance in many econ-
to slower, albeit still-robust, growth—an example
omies and concerns about a “new mediocre” (Lagarde
of a more general phenomenon of changes in trend
2014, 2016). They may also hold clues to the disen-
growth rates documented by Pritchett and Summers
chantment with existing institutions and establishment
(2013). Given these developments (and possibly similar
political parties, and the growing appeal of protection-
underlying shifts over this period in trend growth rates
ism (Lipton 2018).
in other countries), comparisons of current GDP with
precrisis outcomes must be careful to avoid attributing
all of the observed changes to the 2008 crisis.4
2All annexes are available online at www.imf.org/en/Publications/ 5The correlation coefficient between GDP deviations for 2011–13
WEO. and 2015–17 is about 0.90. As shown in Online Annex Figure 2.2.4,
3Online Annex 2.2.B discusses the differences between the chap- the output deviations close to a decade after the 2008 crisis are more
ter’s approach and the standard filtering approach used for separating skewed toward losses than those registered at a similar interval after
output into trend and business cycle components. the 1982 global recession.
4For the United States, for example, there is a range of estimates 6Employment losses are measured as the gap between the number
regarding the postcrisis output loss due to the 2008 financial crisis of employed workers and the number consistent with employment
versus those related to changes in potential output growth already growing at the same rate during the postcrisis period as the economi-
underway prior to the crisis (see CBO 2014; Hall 2014; and cally active cohort between the ages of 15 and 65 (Schanzenbach and
Barnichon, Matthes, and Ziegenbein 2018). others 2017; see Online Annex 2.2.B).

Figure 2.2. Postcrisis Change in Inequality Figure 2.3. Postcrisis Output Deviations from Precrisis Trend,
2015–17
Economies with larger output and employment losses in the initial aftermath of the (Kernel density)
crisis registered greater increases in income inequality compared with the
precrisis average.
Output losses are persistent for a variety of economies, not just those that suffered
a systemic banking crisis in 2007–08.
8 1. Output Deviations
Average change in Gini coefficient
6 Banking crisis No banking crisis

0.05
4
2
0 0.04
–2
–4
–6 0.03
–8
–30 –20 –10 0 10 20 30
Percent deviations from precrisis trend
0.02
8 2. Employment Deviations
Average change in Gini coefficient
6 0.01
4
2
0 0.00
–40 –30 –20 –10 0 10 20
–2
–4 Sources: Laeven and Valencia (2013); and IMF staff calculations.
–6 Note: Distribution of average percent deviations from precrisis trend, 2015–17.
See Online Annex Table 2.2.1 for banking crises country list.
–8
–20 –10 0 10 20
Percent deviations from precrisis trend
Sources: Standardized World Income Inequality Database (Solt 2016); and IMF for the list). Figure 2.3 summarizes the distribution of
staff calculations.
Note: The Gini coefficient is based on income before taxes and transfers and postcrisis output deviations from precrisis trends when
ranges from 0 to 100. The change in Gini coefficient is calculated as the difference deviations are averaged over 2015–17.
between the averages during 2005–08 and 2014–15. Movement from left to right
on the x-axis indicates less negative/more positive average deviations from
Among the 24 economies in the banking crisis
precrisis trend in 2011–13. group, about 85 percent still show negative devi-
ations from the pre-2009 trend a decade after the
2008 meltdown. In light of earlier evidence (see,
Output Remains below Precrisis Trend in More than for example, Abiad and others 2009; Chapter 4 of
60 Percent of Economies the April 2009 WEO; and Blanchard, Cerutti, and
Summers 2015), it is not surprising that economies
The deviations from pre-2009 trends are estimated
in the banking crisis group suffered persistent losses
for two broad samples of economies: those that experi-
thereafter. As Blanchard, Cerutti, and Summers
enced banking crises in 2007–08 (as defined in Laeven
(2015) show, recessions associated with financial
and Valencia 2013) and all other economies.7 Accord-
crises are more likely to lead to persistent shortfalls
ing to the Laeven-Valencia definition, there were bank-
in output relative to precrisis trends. Less credit
ing crises in 24 countries during 2007–08, 18 of which
intermediation—from a combination of supply and
were in advanced economies (see Online Annex 2.2.A
demand factors—is a significant channel (Bernanke
2018). On the supply side, impaired financial systems
7The Laeven-Valencia (2013) definition of a banking crisis is based cannot intermediate credit to the same extent as
on two criteria: significant financial distress (including bank runs before the crash, and postcrisis regulatory tightening
and liquidations) and significant government intervention in the
banking system (including recapitalization, liability guarantees, and
can also affect loan origination. In parallel with the
nationalization). supply disruptions, several factors may have held back

credit demand. These include weak growth expec- Figure 2.4. Postcrisis Output Deviations from Precrisis Trend
tations, impaired corporate and household balance by Country Group, 2015–17
(Kernel density)
sheets weighing on collateral quality, and an impera-
tive to rebuild net worth.
Postcrisis output deviations tend to be large across advanced economies,
However, Figure 2.3 shows the persistence of output emerging markets, and low-income developing countries, with relatively more
losses relative to precrisis trends for several econo- balanced gains and losses for noncommodity-exporting low-income developing
countries and emerging markets than for the other two groups.
mies, not just those that suffered a banking crisis in
2007–08 (consistent with Cerra and Saxena 2017 and AEs LIDC commodity exporters
Aslam and others, forthcoming, who find persistent EMs LIDC noncommodity exporters
0.05
losses associated with most recessions, not just those
associated with financial crises). In the group without
a banking crisis in 2007–08, output remains below 0.04
precrisis trends in about 60 percent of economies. A
possible channel—discussed later in the chapter—that
affected this group is weaker external demand from 0.03
trading partners that suffered banking crises, which
contributed to lower investment and associated capital
shortfalls (also see Candelon and others 2018). 0.02
Grouping the sample by advanced economies,
emerging markets, and low-income developing coun-
0.01
tries shows that output deviations tend to be large
across all groups (Figure 2.4). Output deviations are
relatively more balanced across gains and losses for
0.00
noncommodity-exporting (diversified) low-income –40 –30 –20 –10 0 10 20 30 40
developing countries and emerging market economies
than for the other two groups. More generally, the Source: IMF staff calculations.
Note: Distribution of average percent deviations from precrisis trend, 2015–17.
greater variability in output deviations across emerging AEs = advanced economies; EMs = emerging markets; LIDC = low-income
markets and low-income developing countries com- developing country. See Online Annex 2.1 for country groupings.
pared with advanced economies may reflect the variety
of forces acting on their growth processes, including
commodity price developments, export links to China, with weaker aggregate investment, as documented in
and receipt of outward investment from China (see Chapter 4 of the April 2015 WEO.9
also Aslam and others, forthcoming). Investment shortfalls may have resulted from a
lack of access to credit after the crisis, or from weak
expectations of future growth and profitability (the
Proximate Causes: Sluggish Investment, Capital, and latter view reprises the 1930s notion of secular
Total Factor Productivity Shortfalls stagnation—see Summers 2016 for a discussion; see
The persistence of output deviations suggests also Kozlowski, Veldkamp, and Venkateswaran 2017).
supply-side shifts in the factors of production. As A similar calculation for output, as described earlier in
shown in Online Annex Figure 2.2.3, deviations in this chapter, suggests shortfalls in investment relative
output per worker trace similar patterns to deviations to precrisis trends. Figure 2.5 shows the average across
in aggregate output, indicating that changes in labor all economies of deviations relative to precrisis trends.
input cannot account for the bulk of the observed By 2017, on average, investment was about 25 percent
output deviations.8 This similarity suggests shifts in below precrisis trend.
other factors of production associated, for instance,
9An important exception is China, where the investment share
of GDP rose from below 40 percent in precrisis years to almost

8Nevertheless, as noted in Box 2.1, postcrisis economic perfor- 50 percent after the crisis, driven by credit-fueled expansion of
mance appears to have had an impact on migration and fertility infrastructure, residential and commercial real estate, and corporate
decisions, with attendant implications for future labor input. capital expenditure.

Figure 2.5. Postcrisis Investment Deviations from Precrisis Figure 2.6. Postcrisis Capital Stock Deviations from Precrisis
Trend: Mean Trajectory Trend, 2015–17
(Percent) (Kernel density)
Investment dropped below precrisis trend during the crisis and deviated further in Close to 80 percent of economies that suffered a banking crisis in 2007–08
2012. By 2017, on average, investment was about 25 percent below precrisis experienced shortfalls in capital relative to precrisis trend. Among economies that
trend. did not suffer a banking crisis in 2007–08, about 65 percent appear to be
operating with capital stocks below precrisis trend.
Log investment Trend log investment
40 Banking crisis No banking crisis
0.04
20
0.03
0
–20
0.02
–40
0.01
–60
–80
2000 02 04 06 08 10 12 14 16 17 0.00
–40 –30 –20 –10 0 10 20 30 40
Note: 2008 log investment normalized to zero. Sources: Laeven and Valencia (2013); and IMF staff calculations.
Note: Distribution of average percent deviations from precrisis trend, 2015–17.
See Online Annex Table 2.2.1 for banking crises country list.
Two important consequences of sluggish invest-

ment that may hold clues to why the recovery path.10 At the sectoral level, these capital shortfalls
appears to have been so slow, are shortfalls in the are widespread, extending beyond the construction
capital stock and, to the extent technology is embed- sector, which underwent a needed correction after the
ded in machinery, slower technology adoption. A precrisis boom (Online Annex Figure 2.2.5).
useful way to see this is to decompose the deviations A second possible consequence of sluggish investment
in output per worker from precrisis trends into devi- is slow technology adoption—to the extent that new
ations in capital stock per worker and residual TFP technologies are embodied in equipment. The growth
deviations. A caveat here is that, even though TFP, in accounting approach attributes a significant role to the
principle, reflects both technology and the efficiency residual (TFP) component of deviations from precri-
of combining inputs, in practice it also reflects sis trend in output per worker once the influence of
measurement error in the factors of production deviations in capital per worker is taken into account
and changes in capacity utilization. Evidence from (Figure 2.7). These estimated deviations in TFP from
standard growth accounting techniques (described precrisis trends are consistent with evidence of wide-
in Online Annex 2.2.B and summarized in Fig- spread postcrisis deceleration in TFP growth discussed in
ure 2.6) suggests that there are large capital shortfalls Adler and others (2017). As reported in Table 2.1, the
relative to precrisis trends. Close to 80 percent of median share of output per worker deviation accounted
economies that suffered a banking crisis in 2007–08 for by TFP deviation is close to 80 percent for both
experienced shortfalls in capital relative to precrisis groups of economies. While the evidence points to the
trends. Among economies without a banking crisis in
2007–08, capital stocks of about 65 percent appear 10Online Annex 2.2.B shows that the distributions of capital stock
to be lower than they would be if capital accumu- deviations are not distinguishable across the two groups in a statisti-
lation had followed the extrapolated precrisis trend cal sense, while those of output and TFP are.

Figure 2.7. Postcrisis Total Factor Productivity Deviations Table 2.1. Total Factor Productivity Deviations
from Precrisis Trend, 2015–17 Account for a Large Share of GDP per Worker
(Kernel density) Deviations
(Percent)
Estimated deviations in TFP from precrisis trend are consistent with the evidence Median Share of GDP Deviation Accounted for by Deviation
of a widespread postcrisis deceleration in TFP growth. These TFP deviations in GDP per Worker, 2015–17
account for close to 80 percent of output per worker deviations for both groups of
economies, that is, those that suffered banking crises in 2007–08 and those that Countries without banking crisis in 2007–08 70.4
did not. 2007–08 banking crisis countries 80.5
Median Share of GDP per Worker Deviation Accounted for
Banking crisis No banking crisis by Total Factor Productivity, 2015–17
0.06 Countries without banking crisis in 2007–08 79.3
2007–08 banking crisis countries 78.2
0.05
Note: See Online Annex Table 2.2.1 for banking crises country list.
0.04
Further confirmation of slower innovation and
0.03 technology adoption among countries hit harder by the
crisis is seen through the example of industrial robots—
an observable and much-discussed class of automation
0.02
technology expected to replace human labor in an
increasing range of tasks. (Box 2.2 examines the postcri-
0.01 sis employment impact of industrial robots.)11
An inspection of the industrial robot data (Figure 2.9)
0.00 indicates that the average change in density—measured
–40 –30 –20 –10 0 10 20 30 40 as robot shipments per thousand hours worked—during
the postcrisis period was higher in countries that had
Sources: Laeven and Valencia (2013); and IMF staff calculations.
Note: Distribution of average percent deviations from precrisis trend, 2015–17. smaller postcrisis losses in output.
TFP = total factor productivity. See Online Annex Table 2.2.1 for banking crises As with the general measure of innovation (research
country list.
and development expenditure), the gap in changes in
robot density between high- and low-output-loss coun-
tries is higher among advanced economies than among
importance of TFP deviations in accounting for output
emerging markets. As part of the generalized slower
per worker deviations, the cross-country data do not
investment in the postcrisis period, robot adoption
permit a further separation of TFP deviations into those
may have been affected more negatively in countries
due to sluggish investment from those related to worsen-
hit harder by the crisis.12 This “suppressed-investment”
ing efficiency or other factors unrelated to investment.
Slower Technology Adoption

11As described in Online Annex 2.3.A, data from the International
The estimates of TFP deviation suggest that the
Federation of Robotics, which compiles information on worldwide
pace of technology adoption (and associated pace of shipment of robots, are used to examine the postcrisis diffusion of
upgrading of capital stock with embodied technology) automation technology. The data are reported at the level of indus-
may have slowed following the crisis. However, as tries for 75 countries extending back to 2004 (for some countries,
data are available going back to 1993).
noted above, TFP is an imperfect proxy for the pace of 12While there is possibly an element of reverse causality in these
technology adoption. A clearer picture emerges from correlations (lower robot investment contributed to higher output
examining variables directly associated with innovation loss), empirically, the magnitude of robot investment compared with
manufacturing output in the United States, for example, suggests
and technology adoption. Cross-country evidence on
that the effect of robot investment on manufacturing—as well as
a key innovation input—research and development aggregate—output is small. Based on US Bureau of Economic
spending—suggests that countries with above‑median Analysis data, the International Federation of Robotics (the data
output losses registered slower increases in research and source for robots used in the analysis) reports that the value of
industrial robot shipments to the United States as a share of US
development shares of GDP. This is especially evident gross manufacturing output ranged between 0.016 percent in 2002
among advanced economies (Figure 2.8). and 0.027 percent in 2016.

Figure 2.8. Changes in Research and Development Figure 2.9. Average Change in Robot Density, by Output
Expenditure, by Output Losses and Country Groups Losses and Country Groups, 2010–14
(Percent of GDP) (Robot shipment per 1,000 hours worked)
Countries with above-median output losses registered slower increases in The gap in changes in robot density between high and low loss countries is higher
research and development expenditure shares of GDP. This was especially evident among advanced economies than among emerging markets.
among advanced economies.
0.04
0.3
0.03
0.2
0.02
0.1
0.01
0.0
0.00
–0.1
–0.01
–0.2
–0.02
–0.3 High loss+ Low loss+ High loss Low loss
High loss Low loss High loss Low loss Advanced economies Emerging markets
Advanced economies Emerging markets
Sources: International Federation of Robotics; World Input-Output Database; and

Sources: World Bank, World Development Indicators database; and IMF staff IMF staff calculations.
calculations. Note: Robot density is defined as robot shipment/1,000 hours worked. The bar
Note: The bars depict the difference in averages between 2014–16 and 2011–13. chart shows the interquartile range, and lines display lesser of the maximum
The bar chart shows the interquartile range, and lines display lesser of the (minimum) and +/– 1.5 times the upper (lower) quartile range. High (low) loss
maximum (minimum) and +/– 1.5 times the upper (lower) quartile range. High indicates above (below) median losses in output relative to precrisis trend as
(low) loss indicates above (below) median losses in output relative to precrisis calculated in Online Annex 2.2.B.
trend as calculated in Online Annex 2.2.B. +
denotes differences in medians between high- and low-output loss samples
among advanced economies statistically significant at 10 percent. See Online
Annex 2.3 and Online Annex Table 2.3.2 for further details on data and estimation.
effect likely more than offset any tendency to automate in the buildup to the meltdown of 2008, policy choices
rather than rehire unemployed workers.13 in the immediate aftermath of the crisis, and structural
aspects may have also helped shape postcrisis variation
Policy Frameworks, Measures, and Postcrisis in output performance—in the first instance, by influ-
Output Performance encing countries’ vulnerability to the disruptive forces
the financial meltdown of 2008 unleashed, and subse-
A large number of economies registered output losses quently, by affecting the damage they experienced and
relative to precrisis trends, but the postcrisis experience their ability to recover.
varied by individual country. In part, this variation may Identifying why economies’ responses differed
reflect differences in the nature of the shock at the level can provide important lessons for the most effective
of individual countries. Some suffered severe banking policy responses. The exercise can also help shed light
crises as part of the global financial panic, while others on actions that may help limit damage and facilitate
were affected mostly through their trade and financial recovery in future downturns.
links to the first set of countries. But initial conditions
13Analysis at the industry-country level (Online Annex 2.3.B)

Empirical Approach
corroborates this finding. Industries in advanced economies that
suffered relatively bigger investment and TFP losses during the crisis The previous section noted the persistence of
experienced slower robot diffusion. output losses, with a strong correlation between GDP

deviations for 2011–13 and 2015–17. Understanding and central and eastern Europe, triggering a wave
the sources of variation in output performance during of defaults by overextended property developers and
2011–13 can therefore provide insight into output households unable to roll over their loans, which
patterns observed during 2015–17. further strained the balance sheets of European banks
As explained in Online Annex 2.2.C, the empirical already caught in the web of losses on US subprime
approach estimates cross-sectional regressions similar mortgage exposures. In the euro area, a debilitating
to those of other studies that have examined various nexus soon emerged between banks and sovereigns:
aspects of cross-country variation in the impact of taxpayer bailouts and guarantees of distressed banks
the global financial crisis (Blanchard, Faruqee, and severely undermined public debt sustainability in some
Das 2010; Claessens and others 2010; Lane and countries; in others, weak fiscal positions and widening
Milesi-Ferretti 2010, 2014; Giannone, Lenza, and government spreads critically compromised banks with
Reichlin 2011; Berkmen and others 2012; Tsangarides large holdings of sovereign securities.
2012; Cerra, Panizza, and Saxena 2013). The approach For economies that experienced banking crises
builds on Chapter 4 of the October 2009 WEO, in 2007–08, the loss of intermediation services and
which studies the determinants of medium-term diminished credit volumes, not surprisingly, had a
output losses following financial crises in advanced, far-reaching impact on activity. The associated corpo-
emerging market, and developing economies during rate failures and employment losses undermined the
1970–2002 (see also Abiad and others 2009). ability of borrowers to service their loans, spiraled back
to sap bank balance sheets, forced banks to retrench
credit further, and amplified the output decline.15
The Nature of the Shock Matters The analysis suggests that, on average, countries that
Although the 2008 financial crisis originated in the experienced banking crises suffered a 4 percentage
United States and Europe, it had a global macroeco- point higher output loss during 2011–13 relative to
nomic impact. The origins of the crisis are by now the precrisis trend than those that did not experience
well documented.14 Four aspects are common to most banking crises in 2007–08. (Online Annex Table 2.2.5;
accounts. First, abundant global liquidity enabled a Table 2.2 summarizes the direction of impacts for the
lending boom in the United States, United Kingdom, various drivers.)
euro area, and central and eastern Europe before 2008.
As discussed in Chapter 2 of the October 2018 Global
Financial Stability Report (GFSR), the credit expansion Macroeconomic Imbalances and Financial Factors
was intermediated through complex links between Regardless of whether a country suffered a banking
traditional banks and nonbank financial institutions crisis in 2007–08, tighter financial conditions after the
beyond the regulatory perimeter. Second, as a wave crisis brought out the central role of precrisis financial
of US adjustable rate mortgages began to reset in vulnerabilities in influencing postcrisis output perfor-
2006–07 and subprime borrowers found it difficult to mance. This influence is reflected, at a general level, in
stay current on their loans or refinance them, the US the variation of output performance as a function of
housing market began to turn in an unprecedented, initial macroeconomic and financial imbalances. It is
synchronized manner across many states. Third, unlike also seen in the role played by specific factors, such as
the late-1990s US subprime mortgage collapse, which the pace of precrisis credit growth.
affected mostly loan originators, the financial losses A useful summary statistic of macroeconomic imbal-
were amplified in 2007–08 by the poorly monitored ances is the gap between the actual current account
practice of securitizing subprime loans into complex balance and its level consistent with medium-term
financial products that became impossible to price in fundamentals. This gap can be thought of as a
a declining market. Fourth, tightening global finan- real-time estimate of imbalances resulting from private
cial conditions during 2007–08 hastened the end of
the lending boom in the euro area, United Kingdom, 15Gertler and Gilchrist (2018) examine the relative contribu-
tions of banking disruption and household balance sheets to the

contraction of US employment during the Great Recession. They
14See,for example, Obstfeld and Rogoff 2009; Sorkin 2009; Lewis find that banking disruption is key to the aggregate decline in US
2010; Lowenstein 2010; Rajan 2010; Blinder 2013; Paulson 2013; employment, while household balance sheet strength is relatively
Geithner 2014; Bernanke 2015; Bayoumi 2017; and Toloui 2018. more important for explaining regional variation.

Table 2.2. Impact of Precrisis Conditions on 2011–13 GDP Deviations from Precrisis Trend
(1) (2) (3) (4) (5) (6)
All Countries AEs EMs
Domestic Credit Growth –** –*** –*** –*** –*** –**
Demand Exposure to Advanced Economies –*** – + + – –
Demand Exposure to China + + + +* +** +
Financial Openness –* – – – – –
CA Balance + +*** –
CA Gap +*** +*** +
Share of Manufacturing in GDP + + +
Difficulty of Dismissal –** –* –**
Precrisis GG Debt Change –*** –*** –***
De Facto Peg Dummy –** –*** –
Banking Crisis –** –
Note: + denotes positive impact, – denotes negative impact. Precrisis conditions are averaged over 2005–08. Results in columns (1) and (2) are reported in
Online Annex Table 2.2.5. Results in columns (3) through (6) are reported in Online Annex Table 2.2.7. AEs = advanced economies; CA = current account;
CA Gap = excess external balance, Lee and others (2008); EMs = emerging markets; GG = general government.
*** p < 0.01, ** p < 0.05, * p < 0.1.
and public saving-investment disparities (see Lee and may indicate reluctance on the part of firms during the
others 2008; and Lane and Milesi-Ferretti 2010). The postcrisis recovery phase to expand operations and lock
results suggest that countries with current account themselves into costly contracts in economies where
balances weaker than the level consistent with funda- subsequent exit would be more difficult.
mentals entering the crisis suffered bigger output losses
relative to precrisis trends (Online Annex Table 2.2.5; Spillovers
Table 2.2). This may, in part, reflect the more severe
The results in Table 2.2 are also consistent with
adjustment forced on countries with higher precrisis
spillover effects through trade. Controlling for the
excess deficits.
effect of banking crises, economies relatively more
In addition, countries more dependent on credit
exposed to demand from advanced economies suffered
(those with faster credit growth in the buildup to
larger output losses in the aftermath.
the crisis) suffered larger losses in an environment of
The size of gross external financial exposure acted as
tighter financial conditions.
another key channel through which financial distress
from the crippled core of advanced economies trans-
Labor Market Structure mitted to the rest of the global economy. Countries
Some economies are more flexible than others when more integrated into global financial markets (repre-
it comes to relocating workers in the face of shocks. The sented by larger fractions of external assets and liabil-
strength of employment protection legislation—the bal- ities relative to GDP) experienced bigger deviations
ance it provides between security for workers and flexi- from the precrisis trend.17 This may reflect, in part,
bility for firms—is a key influence on firms’ decisions to retrenchment in global banking after the crisis.
hire new workers. The evidence suggests that economies
in which it was more difficult for firms to terminate Co-operation and Development’s (OECD’s) strength of employment
labor contracts (proxied by an index of ease of dismissal protection indices. The index correlates well with the OECD mea-
compiled by the Centre for Business Research [CBR] at sures for countries covered by the OECD’s indices, as well as with a
typical measure of labor market churn and dynamism (the probabil-
Cambridge University) suffered larger postcrisis losses ity of entering and exiting employment), which can be constructed
in output relative to precrisis trends (Table 2.2).16 This for a limited set of countries along the lines of Elsby, Hobijn, and
Sahin (2013), as described in Online Annex 2.2.C.
17This is consistent with Perri and Quadrini (2018), who develop a
16The Cambridge University CBR index (Adams, Bishop, and model of global, synchronized recessions that follow from cross-border
Deakin 2016) is based on an average of nine detailed indicators of transmission of liquidity shortages in highly integrated capital mar-
dismissal procedures constructed using leximetric coding methodol- kets. The extensive cross-border financial links—particularly among
ogy on country-level labor legislation. The index is used here because advanced economies—on the eve of the crisis was unprecedented and
it has broader country coverage than the Organisation for Economic may have compounded countries’ vulnerabilities. See also Chapter 4 of

There is a similar pattern for postcrisis investment Figure 2.10. Probability of Banking Crisis
deviations among countries that did not experience a (Probability)
banking crisis in 2007–08 (Online Annex Table 2.2.6).
Stronger restrictions in 2006 on banks’ ability to underwrite, broker, and deal in
In particular, countries with stronger trade ties to securities; offer mutual fund products; and engage in insurance underwriting, real
advanced economies going into the crisis experienced estate investment, development, and management are associated with a lower
probability of banking crisis in 2007–08.
larger deviations in investment during 2011–13
relative to precrisis trends. This finding is consistent
0.8
with the earlier observation (Figure 2.6) that persistent Probit
Logit
capital shortfalls were observed also in countries that Linear probability model
did not experience a banking crisis in 2007–08.
An important offsetting influence on weak demand 0.6
from advanced economies during this period was
demand from China. China’s 4 trillion yuan stimulus
during 2008–11 (close to 10 percent of 2008 GDP) 0.4
supported a large nationwide infrastructure expansion
and construction of social housing, with associated
favorable impacts on exporters of commodities and
heavy equipment (Ahuja and Nabar 2012). The results 0.2
in Online Annex Table 2.2.7 (summarized in Table 2.2),
grouped according to advanced and emerging mar-
ket economies, indicate that economies whose export 0.0
baskets were more exposed to China before the crisis –2 –1 0 1 2
Strength of restrictions on banking activities
benefited disproportionately in the aftermath from
higher exposure to China’s domestic demand (measured
Sources: Barth, Caprio, and Levine (2013); and IMF staff calculations.
as the share of trading partner demand accounted for by Note: Movement from left to right on the x-axis indicates stronger restrictions on
China), especially among emerging market economies. banking activities. Figure is based on Online Annex Table 2.2.3.
Precrisis Policies and Policy Frameworks overall intensity of financial sector monitoring activity;
The incidence of bank crises in 2007–08 was a key the porosity of the regulatory perimeter and opportuni-
driver of subsequent losses. Regulatory and supervisory ties for regulatory arbitrage) likely also played a role.
structures may thus have played a preemptive role in In general, the initial policy space available prior
influencing subsequent damage. The bank regulation to a crisis can affect the extent of activity decline
index constructed by Barth, Caprio, and Levine (2013) afterward (Blanchard, Dell’Ariccia, and Mauro 2010;
illustrates this link. Specifically, stronger restrictions in Jordà, Schularick, and Taylor 2016; Romer and Romer
2006 on banks’ ability to underwrite, broker, and deal 2018). For the 2008 episode specifically, countries
in securities; offer mutual fund products; and engage in with smaller increases in general government debt
insurance underwriting, real estate investment, devel- over 2005–08 experienced smaller losses relative to
opment, and management are associated with a lower trends (Table 2.2). Countries with lower public sector
probability of a banking crisis during 2007–08 (Fig- borrowing requirements going into the crisis appear
ure 2.10).18 However, the index measures the strength to have had more room to deploy fiscal policy for
of restrictions only on specific aspects of bank activity. demand support in the immediate aftermath.
Other dimensions (for instance, strength of capital, Policy frameworks also appear to matter for postcri-
funding, and liquidity requirements; the accompanying sis output outcomes. Exchange rate flexibility is associ-
supervisory approach to stress-testing balance sheets; ated with less damage, pointing to a buffering role of
nominal exchange rates (Table 2.2). This finding may,
the April 2009 WEO, which documents the role of international links in part, reflect the difficulties experienced by some
in transmitting financial stress across borders. euro area economies. In these countries, the absence
18The association shown here is robust to controlling for some
other influences on the likelihood of a bank crisis (Online Annex of an independent nominal exchange rate, together
Table 2.2.4). with fiscal stress and the lack of a common area-wide

Figure 2.11. Postcrisis Deviations of Euro Area and Box 2.3), and—despite substantial progress toward a
Other Advanced Economies banking union and the creation of the European Sta-
(Percent)
bility Mechanism for crisis management—remaining
gaps in the euro area architecture.19
The median and PPP GDP-weighted mean of output loss for euro area economies
are higher than for other advanced economies.
Euro area Other advanced economies Extraordinary Actions Taken in the Aftermath
of the Crisis
1. Median
0 Several countries took exceptional and unprecedented
policy measures to support their economies after the
–2
2008 financial crisis. In many cases, notably among the
–4 advanced economies most severely affected by the crisis,
–6 the measures comprised (1) central bank monetary pol-
icy actions—unconventional monetary policy support
–8
through asset purchases as policy rates approached their
–10 effective lower bounds, and liquidity support to specific
–12
segments of credit markets through targeted central
2011–13 2015–17 bank facilities; (2) discretionary fiscal stimulus; and (3)
financial sector operations—bank balance sheet stress
2. PPP GDP-Weighted Mean
0
tests, government guarantees of banking sector liabil-
ities, purchases of toxic assets from banks, and capital
–2 injections. Central banks also established ad hoc bilat-
–4 eral swap lines to support foreign exchange liquidity in
jurisdictions beyond home markets.
–6
Advanced economy monetary policy actions, in
–8 particular, represented a significant change in the
–10 approach to providing monetary accommodation—
necessitated in some cases by central banks rapidly
–12
2011–13 2015–17 reducing policy rates to their effective lower bounds
during the crisis (Bernanke 2017). The particular mix
Source: IMF staff calculations. of tools varied across individual cases, but generally
Note: Other advanced economies are advanced economies that are not in the euro included a combination of quantitative easing (mas-
area. PPP = purchasing power parity.
sive balance sheet expansion with purchases mainly of
government bonds, mortgage-backed securities, and
corporate bonds); state-dependent forward guidance
banking union and fiscal backstop, meant the burden (specifying particular levels of unemployment and
of adjustment after the crisis fell entirely on domestic inflation as conditions for rate hikes); negative interest
prices and output. rates (charging commercial banks a penalty on excess
The median output loss for euro area economies is reserves held at the central bank); and yield-curve
notably higher than for other advanced economies in control (targeting the yields of longer-maturity govern-
2011–13 (Figure 2.11), covering an intense phase of ment bonds through central bank purchases).
the sovereign debt crisis, deposit flight from stressed Estimates of the impact of advanced economy
euro area economies, and financial fragmentation central banks’ quantitative easing on interest rates and
within the euro area (see IMF 2012, 2013a). The dif- financial conditions vary (Gagnon 2016). In general,
ference in losses widened through 2015–17, pointing the positive effect of the actions on domestic output in
to a weaker recovery compared with other advanced
economies. The divergence may, in part, reflect the
19Thomsen (2017); Arnold and others (2018); and Berger,
limited policy levers available within a currency union
Dell’Ariccia, and Obstfeld (2018) discuss the reforms implemented
for adjustment to asymmetric shocks, differences in to strengthen the euro area architecture and the remaining steps to
the speed of financial sector repair (as discussed in complete the banking and fiscal union.

Table 2.3. Financial Sector Support and Discretionary Fiscal Stimulus in Group of Twenty Economies
(Percent of GDP)
1. Headline Support for the Financial Sector (as of February 2009)
Purchase of Central Bank Central Bank
Capital Assets, Lending Support with Liquidity
Injection by Treasury Treasury Backing Support Guarantees Total
(A) (B) (C) (D) (E) (A+B+C+D+E)
G20 Average (PPP GDP-weighted) 2.0 3.3 1.0 9.2 14.3 29.8
Advanced Economies 2.9 5.0 1.2 12.9 21.3 43.3
Advanced Europe 2.4 3.6 2.1 1.0 19.5 28.6
Emerging Markets 0.3 0.1 0.3 1.8 0.2 2.7
2. Crisis-Related Discretionary Fiscal Stimulus in G20 Economies (as of October 2010)

2009 2010 2011
G20 Average 2.1 2.1 1.1
Advanced Economies 1.9 2.1 1.2
Emerging Markets 2.4 2.0 0.9
Sources: IMF (2009); IMF Fiscal Affairs and Monetary and Capital Markets departments database on public interventions; Chapter 1 of the November 2010
Fiscal Monitor.
Note: Panel 1 is calculated based on country statistics originally published in IMF (2009). The data on guarantees for Australia are based on Schwartz and Tan
(2016). In panel 1, G20 calculations do not include Mexico and South Africa. G20 = Group of Twenty.
advanced economies and imports from trading partners instances of asset purchase programs by advanced
is believed to have outweighed negative effects as a economy central banks and therefore more easily stud-
result of elevated capital inflows and currency appre- ied in a regression framework to assess their impact on
ciation pressure elsewhere (IMF 2014). More broadly, output deviations.
quantitative easing may have also helped stabilize activ- Estimating the immediate effect of the actions is
ity by reducing the tail risk of debilitating asset price difficult. In the case of discretionary fiscal stimulus,
declines. Nevertheless, the actions were the subject of for example, causality runs in both directions, with
controversy, with policymakers in emerging market larger output collapses likely to prompt larger policy
and developing economies, at times, raising concern responses, all else equal. It is nonetheless possible to
about adverse spillovers from advanced economy cen- detect lagged effects of the measures on output devia-
tral banks’ unconventional monetary policy approaches tions from precrisis trends averaged over 2015–17.
(Mantega 2010; Zhou 2010; Rajan 2014). As shown in Figure 2.12, conditional on the size of
The analysis in this chapter focuses on the impact initial losses during 2011–13, quasi-fiscal actions taken
of fiscal and quasi-fiscal measures in support of the to stabilize the financial sector helped limit damage
financial sector undertaken by some economies in during 2015–17. Overall headline support for the
the aftermath of the crisis (Table 2.3). The Group financial sector has a statistically significant positive
of Twenty (G20) economies, for example, on aver- correlation with subsequent output deviations from
age, injected discretionary fiscal stimulus of just over trend; among the specific actions, capital injections
2 percent of GDP in 2009 and 2010. (The IMF and guarantees appear to have helped limit subsequent
was among the early advocates of the effort in the output losses. These interventions may have helped
days leading up to the November 2008 G20 Sum- thaw credit markets, and resumption of credit services
mit.)20 The number of such actions is larger than the subsequently contributed to raising output.
Beyond action at the national level, as discussed
20During 2008 and 2009, the G20 forum (Argentina, Australia, in Chapter 2 of the October 2018 GFSR, there were
Brazil, Canada, China, France, Germany, India, Indonesia, Italy, extensive multilateral efforts to strengthen financial
Japan, Mexico, Russia, Saudi Arabia, South Africa, South Korea,
regulatory standards (aimed at expanding the regu-
Turkey, United Kingdom, United States, European Union) was piv-
otal in forging international consensus on fiscal expansion, augment- latory perimeter, containing the buildup of systemic
ing the lending resources of the IMF and multilateral development risk, strengthening resilience to shocks, and develop-
banks, and the need to strengthen financial regulation (see https:// ing resolution frameworks). Multilateral cooperation
www.g20.org/en/g20/timeline). For the IMF’s November 2008 call
for fiscal stimulus by the G20 economies, see http://www.imf.org/en/ also helped craft an important component of the
News/Articles/2015/09/14/01/49/pr08278. monetary response to the crisis, with the IMF pro-

Figure 2.12. Impact on 2015–17 GDP Deviations from Summary

One Standard Deviation Increase in Drivers
(Percent) The 2008 financial crisis had its roots in the US
housing boom of the preceding half-decade. Its impact
Actions taken to stabilize the financial sector helped limit damages during was seen worldwide from shuttered maquiladora facto-
2015–17. Overall headline support for the financial sector has a statistically ries in Mexico to the restructuring of regional savings
significant positive correlation with subsequent output deviations from precrisis
trend. Among specific actions, capital injections and guarantees have helped limit and loan cajas in Spain and extended joblessness for
subsequent output losses. migrant workers in China’s Pearl River Delta. Output
losses following the 2008 financial meltdown were per-
sistent and experienced by a broad set of countries, not
Total headline support for just the group afflicted by banking crises at the time.
**
financial and other sectors

Protracted weak investment after the crisis was a major
contributing factor, associated with persistent shortfalls
in capital and total factor productivity, relative to pre-
Capital injections crisis trends, and slower technology adoption among
*
countries hit harder by the crisis.

The crisis prompted a still-ongoing rethink of the
Purchase of assets and nature of economic fluctuations, as well as of the
lending by treasury
role of policy frameworks and measures to combat
downturns. The policy lessons of the crisis discussed
Guarantees in this chapter follow from the lens adopted to view
(excluding deposit its aftermath and to understand why the recovery
*
insurance)
appeared so slow in many countries. Other important
0 1 2 3 4 developments covered in previous WEO reports, such
Percent deviations from precrisis trend as the declining share of labor income (Chapter 3 of
the April 2017 WEO), subdued wage growth, and the
Note: Movement from left to right on the x-axis indicates less negative/more
rise of part-time work (Chapter 2 of the October 2017
positive deviations from precrisis trend. Extraordinary measures were taken during WEO), pose additional policy challenges for ensuring
2008–09. Coefficient bars correspond to estimates in Online Annex Table 2.2.8. the income security and welfare of those who rely
*** p < 0.01, ** p < 0.05, * p < 0.1.
mostly on their labor income.
The evidence documented in this chapter suggests
that policy choices in the run-up to the crisis and in
viding unconditional financial resources to its mem- its immediate aftermath influenced postcrisis out-
bers through a general allocation of SDR 204 billion put performance in multiple ways. Stronger banking
($316 billion) during August–September 2009.21 In regulation—proxied by restrictions on certain aspects
addition, several economies relied on the global finan- of bank activity—appears to have played a preventive
cial safety net to ease their adjustment to the funding role by lowering the probability of a banking crisis in
shock after the crisis. The IMF, for example, approved 2007–08. The finding is relevant for ongoing debates
SDR 420 billion in support to its members during on rolling back the regulatory standards adopted fol-
2008–13, of which SDR 119 billion was drawn lowing the crisis.
during that interval.22 Countries with stronger fiscal positions entering the
crisis suffered smaller losses, suggesting that greater
room for policy maneuver may have helped defend
21The IMF’s special drawing right (SDR), an international reserve
against harm. Extraordinary fiscal and quasi‑fiscal
asset based on a basket comprising the US dollar, Chinese renminbi,
actions to support the financial sector after the crisis
Japanese yen, euro, and British pound, is a claim on freely usable
currencies of IMF members. The 2009 general SDR allocation appear to have helped lessen output losses over the
augmented IMF members’ international reserves, with the aim of medium term. Economies that moved quickly to assess
easing postcrisis liquidity constraints (https://www.imf.org/en/News/ the health of their banking systems and recapitalize
Articles/2015/09/14/01/49/pr09283).
22The gross figure includes precautionary arrangements. See IMF banks appeared to have suffered smaller output losses
(2015) for details. subsequently. As IMF (2013c), Auerbach (2017),

Blanchard and Summers (2017), and Furman (2018) buildup of financial vulnerabilities, as discussed in the
note, there is renewed recognition of discretionary April and October 2018 GFSRs. The large accumu-
fiscal policy as a countercyclical demand management lation of public debt and the erosion of fiscal buffers
tool. Moreover, as the analysis shows, China’s large in many economies following the crisis point to the
fiscal stimulus during 2008–11 appears to have had urgency of rebuilding those defenses to prepare for the
favorable spillovers on trading partners. Altogether, the next downturn. Moreover, some of the crisis manage-
evidence presented here suggests some confirmation ment tools deployed in 2008–09 are no longer available
of the efficacy of fiscal measures in limiting persistent (the Federal Reserve’s bailouts of individual institutions,
losses after a recession. And as noted in earlier IMF for example), suggesting financial rescues in the future
research (IMF 2014), unconventional monetary policy may not be able to follow the same playbook.
actions by advanced economy central banks helped Beyond these aspects, more fundamental challenges
limit output declines and employment losses at home relate to long-lasting legacies of the crisis. There are
while supporting imports from abroad. already signs of possible long-term consequences of
The policy efforts of the past decade helped fore- the crisis on potential growth through its impacts on
stall an even worse outcome with deeper output and migration, fertility, and future labor input (Box 2.1).
employment losses. After faltering at times over the And societal support for openness and global economic
past 10 years, the global economic recovery experi- integration appears to have weakened in many coun-
enced a long-awaited synchronized growth upswing in tries after the crisis. The corollary of these develop-
2017–18. Nevertheless, large challenges loom for the ments is the rising appeal of protectionist nostrums
global economy. The extraordinary policy actions to and populism. A fuller reckoning of such long-lasting
prevent a second Great Depression have had important legacies of the 2008 financial crisis must necessarily
side effects. The extended period of ultralow interest await the broader perspective that will emerge with
rates in advanced economies has contributed to the further passage of time.

Box 2.1. The Global Financial Crisis, Migration, and Fertility

Empirical and anecdotal evidence suggest that Figure 2.1.1. International Migration and the
changes in economic performance affect migration Global Financial Crisis
flows and fertility rates. This box explores the rela-
tionships between postcrisis economic performance, AEs EMs LIDCs
policies, migration, and fertility. The main finding of
the box is that postcrisis economic performance had 4 1. Net Migration Rate
a significant impact on both migration and fertil- (per 1,000 of population)
3
ity. Through these channels, the crisis has likely left
2
long-lasting scars on future growth. The box also iden-
tifies several policies associated with significant impacts 1
on migration and fertility. 0
The Great Recession and Migration –1
The decades leading up to the global finan- –2

cial crisis saw large increases in net migration –3
(immigration-emigration) rates between advanced 1980 85 90 95 2000 05 10 15
economies.1 This trend, however, reversed after the
crisis. Meanwhile, net migration has been consis- 140 2. Output Losses and Emigration
Percent change in emigration rate,
120
tently neutral in emerging markets through both 100
2014–16 versus 2005–08
periods, while low-income developing countries have 80

increased net migration rates in the postcrisis years, 60
40
even as they are generally more prone to volatile net 20
migration rates (Figure 2.1.1, panel 1). Motivated by 0
this heterogeneity of net migration among coun- –20
–40
try groups, the analysis examines the relationship –60
between the changes in trends before and after the –80
crisis, looking at per capita GDP and migration –100
–0.3 –0.2 –0.1 0.0 0.1 0.2 0.3
flows by using data on migration inflows from Source country GDP per capita losses, 2011–13
143 source countries to 20 destination advanced
economies.2 Sources: Organisation for Economic Co-operation and
Immigrants are typically more vulnerable to Development; United Nations Department of Economic and
economic shocks than natives. They are often over- Social Affairs, World Population Prospects: The 2017
Revision; and IMF staff calculations.
represented in sectors most sensitive to the business Note: AEs = advanced economies; EMs = emerging market
cycle (OECD 2009) and may face discrimination in economies; LIDCs = low-income developing countries. Net
a tight labor market (Arai and Vilhelmsson 2004). migration rate by country group is population-weighted average.
Losses are based on calculations in Online Annex 2.2.B.
Immigrants have also responded to changes in labor
demand more strongly than natives (Kahanec and
Guzi 2017). Simple correlations confirm the conjec-
ture that migrants respond to economic performance
(Figure 2.1.1, panel 2), measured by the deviations
The authors of this box are Christopher Johns, Mico Mrkaic, of GDP per capita from precrisis trend (calculated as
and Yuan Zeng. described in Online Annex 2.2.B).
1Net migration rate is defined as the number of immigrants
minus the number of emigrants over a period, divided by the

Beyond the correlations, the analysis explores the
person-years lived by the population of the destination country links between economic performance and migra-
over that period. It is expressed as net number of migrants per tion in a multivariate setting, controlling for the
1,000 population. additional main drivers mentioned in the October
2The analysis uses migration inflows, given that inflows are
2016 World Economic Outlook—structural factors
tracked more precisely and more frequently than bilateral migra-
tion outflows and migrant stocks. Data on bilateral migration
and immigration policies. While the box’s discussion
inflows facilitates accurate analysis of the push and pull factors centers on the role of economic factors in migra-
influencing international migration. tion decisions, it should be mentioned that some

Box 2.1 (continued)

migration decisions are driven entirely by such factors Figure 2.1.2. Impact on Emigration Rate from
as political instability and war in the source country One Standard Deviation Increase in Drivers at
or region. To avoid biasing results, migration flows Different Horizons
data exclude flows of refugees and asylum seekers.3 (Percentage points)
Figure 2.1.2, panel 1 shows the impact of losses in
GDP per capita on differences between emigration Short term Medium term
rates in 2011–13 and 2014–16 compared with years
before the crisis (2005–08).4 Losses in GDP per cap- 1. Effect of GDP per Capita Losses
ita significantly impact migration flows in the short GDP per capita losses (D)
and medium terms. In addition to economic perfor- GDP per capita losses (S)
mance, migration flows are affected by the strength of
Education level (D)
poverty constraints in source countries5 and the GDP
per capita in the destination relative to the source Education level (S)
country, education in destination and source coun- Poverty constraint (S), 2005–08
tries, and the distance between destination country
GDP per capita (D/S), 2005–08
and source country.6,7
Policies imposed in the wake of the crisis to limit Distance
migration and reduce competition in labor markets –20 –10 0 10 20
also affect migration (Figure 2.1.2, panel 2). The
analysis examines restrictions on legal entry, stay, 2. Effect of Policies
and quotas (an increase in each variable denotes GDP per capita losses (D)
greater restrictiveness). Increased postcrisis restric- GDP per capita losses (S)
tions significantly reduced migration flows, mostly Legal entry (D), 2011–13
in the medium term, over and above the impact of
Quotas (D), 2011–13
economic losses.
Poverty constraint (S), 2005–08
The Great Recession and Fertility
GDP per capita (D/S), 2005–08
During a recession, relatively elevated unemploy-
Distance
ment rates may lead to deferred decisions on mar-
riage, having children, or both. In nearly all recent –20 –10 0 10 20
Average percent change from
recessions in advanced economies, the impact on precrisis migration rate
fertility has been mainly to postpone births, which
contributes to a short-run reduction in the number Sources: International Migration Institute; Organisation for
of births in the aftermath (long-run effects tend to be Economic Co-operation and Development; Standardized World
Income Inequality Database (Solt 2016); World Bank, World
Development Indicators database; and IMF staff calculations.
Note: Explanatory variables are contemporaneous with
3Inflows of foreign population data are from the Organisa- dependent variable unless noted otherwise. All postcrisis
tion for Economic Co-Operation and Development (OECD). variables except GDP per capita losses are average changes
Refugees and asylum seekers are excluded from the data for all from precrisis (2005–08) levels. Losses are based on
calculations in Online Annex 2.2.B. All coefficients are
countries except: Germany, Netherlands, and Norway—included
statistically significant at 5 percent. Increases in policy variables
if living in private households (as opposed to reception centers correspond to increases in restrictiveness. S = source country;
or hostels for immigrants); and United Kingdom—included if D = destination country. Short term = 2011–13 average;
stayed in country longer than one year. Medium term = 2014–16 average.
4Emigration rate is defined as inflows to destination country
from source country over a period, divided by 1,000 population

in source country.
5Defined as the disposable income Gini coefficient divided by less pronounced).8 Although immigration may be a
the square of PPP GDP per capita. partial solution for low fertility and an aging popula-
6Distance is defined as great-circle distance between most
tion in the short term, in the long term, immigrants’
populated cities in destination country and source country.
7Controls based on measures used in Borjas (1987); Hatton and
Williamson (2002); and Clark, Hatton, and Williamson (2007). 8Neels (2010); Cherlin, Cumberworth, and Morgan (2013).

Box 2.1 (continued)

Figure 2.1.3. Total Fertility Rate Figure 2.1.4. Impact of Crisis Exposure on
(Number of births per woman) Fertility Rate at Different Horizons
(Average change in fertility rate on x-axis;
OECD AEs EMs United States postcrisis minus precrisis)
Italy Spain Greece
Short term Medium term
2.2
2.0
*
GDP losses
1.8
1.6
*
GDP per capita losses
1.4
1.2
***
Employment losses
***
1.0
2000 02 04 06 08 10 12 14 16
–0.10 –0.05 0.00 0.05 0.10 0.15
Sources: Organisation for Economic Co-operation and
Development (OECD); World Bank, World Development Sources: Organisation for Economic Co-operation and
Indicators database; and IMF staff calculations. Development; and IMF staff calculations.
Note: OECD is the average fertility rate for OECD and partner Note: Explanatory variables are contemporaneous with
countries. AEs = OECD and partner advanced economies; dependent variable. Average changes in fertility rate are the
EMs = OECD and partner emerging market economies. See difference between postcrisis term and precrisis (2005–08) level.
Online Annex 2.1 for country list. Losses are based on calculations in Online Annex 2.2.B. Short
term = 2011–13 average; Medium term = 2015–16 average.
* p < .10; ** p < .05; *** p < .01.
fertility rates generally converge to that of natives

(Espenshade 1994).
In the decade before the crisis, the total fertility rate same time span. These persistently low fertility rates
rose in several advanced economies, only to decline over the past decade may weigh on future labor input
afterward (Figure 2.1.3).9 In the United States, the and thus weaken potential growth in the long run.
rate fell from a peak of 2.12 in 2007 to 1.8 in 2016. Evidence from OECD and partner countries shows
Similarly, the birth rate of foreign-born women (ages that average changes in the fertility rate for the post-
15–50) in the United States declined by 16 births per crisis period relative to the precrisis period (2005–08)
thousand women from its peak of 76 in 2008 to its have been negatively impacted by the crisis through
2016 level. For European countries, such as Greece several channels, of which employment losses were
and Spain that suffered a double-dip recession, the the most significant (Figure 2.1.4). Further evidence
fertility rate decreased from 1.5 to about 1.3 over the in the literature (Sobotka, Skirbekk, and Philipov
2011) shows that other complex social changes (higher
9Total fertility rate in a specific year is defined as the total
female labor participation rate, smaller desired family
number of children that would be born to each woman if she size, and so on) and burdened welfare systems could
were to live to the end of her child-bearing years and give birth
to children in alignment with the prevailing age-specific fertility
affect women’s reproductive decisions.
rates. It is calculated by aggregating age-specific fertility rates as The fertility rate can be affected by labor market
defined over five-year intervals. policies as well. Figure 2.1.5 shows how policies

Box 2.1 (continued)

affected fertility after the crisis. On one hand, the Figure 2.1.5. Impact on Fertility from One
result in panel 1 demonstrates that a higher precrisis Standard Deviation Increase in Drivers at
tax wedge on couples reduces fertility in the short Different Horizons
term. On the other hand, panel 2 suggests that post- (Average change in fertility rate on x-axis; postcrisis
minus precrisis)
crisis increases in family allowances and improvements
in job protection during maternity are associated
with higher fertility rates. These findings are in line Short term Medium term
with evidence and case studies from European Union
1. Effect of Precrisis Policies
countries.10 Public spending on
child education and care
Child-care benefits
Family allowance
Total protected
maternity leave
Average tax wedge
**
of couple
–0.10 –0.05 0.00 0.05 0.10 0.15
2. Effect of Policy Changes

Public spending on
child education and care
Child-care benefits
***
Family allowance
Total protected
**
maternity leave
Average tax wedge
of couple
–0.10 –0.05 0.00 0.05 0.10 0.15
Sources: Organisation for Economic Co-operation and

Development; and IMF staff calculations.
Note: Explanatory variables are contemporaneous with
dependent variable. Precrisis policy variables are average of
period 2005–08. Policy changes are average postcrisis
changes from precrisis (2005–08) levels. Average changes
in fertility rate are of difference between postcrisis term and
precrisis (2005–08) level. Short term = 2011–13 average;
10See, for example, Hoem (2008), Kalwij (2010), and Medium term = 2015–16 average.
Thévenon (2011). * p < .10; ** p < .05; *** p < .01.

Box 2.2. The Employment Impact of Automation Following the Global Financial Crisis: The Case of
Industrial Robots
As discussed in the chapter, an important change in Figure 2.2.1. Effect of Robot Diffusion on
the production process after the global financial crisis Employment Growth
appears to be the pace of technology adoption. This box (Percent)
addresses the following questions related to technology
All
adoption, using the example of industrial robots: How High output loss
did the diffusion of robots affect employment in the Low output loss
aftermath of the crisis? What type of workers were par-
0.04 1. All Countries
ticularly affected? Did certain labor market policies alter
the impact of robot adoption on employment?
Forces of automation were at work prior to the 0.02
crisis (Autor, Levy, and Murnane 2003; Goos and
Manning 2007; Acemoglu and Autor 2011; Autor 0.00
and Dorn 2013), and one much-discussed aspect of
the transformation of the workplace is the diffusion –0.02
of industrial robots. Yet, existing work has mostly
focused on exploring precrisis diffusion of automation ***
–0.04
in the United States (Autor, Levy, and Murnane 2003;
Acemoglu and Autor 2011; Autor and Dorn 2013; 0.04 2. Advanced Economies
Acemoglu and Restrepo 2017), and in a few European
Displacement effect
countries (Graetz and Michaels forthcoming; Chiacchio, 0.02 dominates
Petropoulos, and Pichler 2018). Thus, less is known
about postcrisis robot diffusion in and beyond these 0.00
countries. Exploring these recent developments may
provide some perspective on possible future workplace
–0.02
dynamics and labor market outcomes, where artificial-
intelligence-powered equipment is expected to replace
***
human input in an expanding range of nonroutine –0.04
tasks (Berg, Buffie, and Zanna 2017; Frey and Osborne 1.0 3. Emerging Markets
**
2017; Acemoglu and Restrepo, 2018 and forthcoming). 0.8
0.6
Effect of Robot Diffusion on Employment
0.4
As noted in Acemoglu and Restrepo (2017), robot 0.2
diffusion can affect employment in different ways. 0.0
Greater diffusion of robots can affect employment –0.2
negatively through displacement (by directly replacing –0.4 Productivity effect
workers performing certain tasks), but also positively, –0.6 dominates
through productivity gains, as robots can free up –0.8
human labor for other tasks, incentivize investment,
Sources: IFR (2017); World Input-Output Database; and IMF
and create employment. staff calculations.
Estimation results show that increased robot Note: Robot diffusion is defined as average change in robot
diffusion in industries located in countries with more shipments/1,000 hours worked 2010–14. Error bars around
coefficient estimate are two standard errors. Losses are
negative output losses during the crisis is associated based on calculations in Online Annex 2.2.B. Figure is based
with lower employment growth (Figure 2.2.1) in the on coefficients in Online Annex Table 2.3.4.
* p < .10; ** p < .05; *** p < .01.
The authors of this box are Wenjie Chen and Malhar Nabar.

Box 2.2 (continued)

aftermath of the crisis. This is particularly driven by Figure 2.2.2. Hollowing-Out Effect of Robot
industries in advanced economies with relatively bigger Diffusion on Employment Growth
output deviations relative to precrisis trend. In emerg- (Percent)
ing markets with relatively lower output deviations
High medium-skill share
relative to precrisis trend, increased robot diffusion is Low medium-skill share
associated with higher employment growth.
0.06 1. All Countries
Hollowing Out of the Employment-Skills 0.04
Distribution 0.02
The negative association between labor and robot 0.00
diffusion appears to be more pronounced in indus- –0.02
tries initially more reliant on medium-skilled work- –0.04
ers. The effect is largely seen in advanced economies **
–0.06
(Figure 2.2.2). This finding is consistent with the
–0.08
hollowing-out effects documented by Autor, Levy, and
Murnane (2003), and Goos, Manning, and Salo- –0.10
mons (2014). 0.06 2. Advanced Economies
0.04 Hollowing-out effect
Labor Market Policies
0.02
To explore whether labor market policies can mit- 0.00
igate the impact of robot diffusion on employment,
–0.02
regression analysis is conducted on samples divided by
–0.04 **
the severity of crisis exposure.1
A consistent picture emerges (Figure 2.2.3): the –0.06
postcrisis displacement effect of robots on employ- –0.08
ment was more pronounced in countries with more –0.10
rigid labor market policies and less labor market 1.0 3. Emerging Markets
dynamism (churn) prior to the crisis.2 More specifi-
0.5
cally, lower active labor market spending as a share of
GDP, stricter dismissal policies, less churn in the labor 0.0
market, and more stringent employment protection –0.5
legislation are associated with higher displacement –1.0
effects of robot diffusion in countries that experienced
–1.5
relatively high output losses.
–2.0
–2.5
1Four specific measures of labor market policy are under Sources: IFR (2017); World Input-Output Database (WIOD);
consideration: (1) active labor market policy (ALMP) spending and IMF staff calculations.
as share of GDP, (2) ease of dismissal index by Cambridge Note: Robot diffusion is defined as robot shipment/1,000
University’s Center for Business Research, (3) labor churn hours worked. Level of worker skills is based on education
as calculated in Online Annex 2.2.B, and (4) employment attainment from WIOD. Medium-skilled workers have
protection legislation index compiled by the Organisation for attained secondary and/or postsecondary nontertiary
Economic Co-operation and Development. All measures are education in 2009. Error bars around coefficient estimate
calculated as precrisis averages to capture the initial extent of are two standard errors. Figure is based on coefficients in
Online Annex Table 2.3.5.
labor market rigidities.
2Labor market dynamism, also referred to as job churn, is * p < .10; ** p < .05; *** p < .01.
measured as described in Online Annex 2.2.B, following Elsby,
Hobijn, and Sahin (2013).

Box 2.2 (continued)

Figure 2.2.3. Labor Market Policies and In sum, industries in advanced economies with rel-
Effect of Robot Diffusion on Employment atively bigger output losses experienced displacement
Growth effects from robot diffusion in the aftermath of the
(Average change in employment growth, 2010–14) crisis. This negative effect on employment growth was
particularly severe in industries in advanced economies
ALMP Job churn with relatively large shares of medium-skilled workers.
Dismissal EPL
At the same time, in countries with more rigid labor
market policies and less churn, the labor displace-
0.20 1. High Output Losses
ment effect of robot diffusion was more pronounced,
0.15 suggesting that policies supportive of creating more
0.10 flexible labor markets can help absorb employment
0.05
displacement associated with automation.
0.00
–0.05 *** *** ***
–0.10 **
–0.15
More flexible Less flexible
labor market labor market
0.20 2. Low Output Losses

0.15
0.10
0.05
0.00
–0.05
–0.10
–0.15
More flexible Less flexible
labor market labor market
Sources: Cambridge University Center for Business Research

(CBR); IFR (2017); Organisation for Economic Co-operation
and Development; World Input-Output Database; and IMF
staff calculations.
Note: More flexible labor market comprises countries that
have above-median ALMP spending (percent of GDP),
above-median job churn rates, below-median dismissal
regulations as measured by CBR, and below-median EPL.
Error bars around coefficient estimate are two standard
errors. Losses are based on calculations in Online Annex
2.2.B. Figure is based on coefficients in Online Annex Table
2.3.6. ALMP = active labor market policy; EPL = employment
protection legislation.
* p < .10; ** p < .05; *** p < .01.

Box 2.3. The Role of Financial Sector Repair in the Speed of the Recovery
As the financial crisis started rattling markets, Figure 2.3.1. Containment and Resolution
policymakers broadly followed the crisis management
rulebook: step one—stop panic from spreading (con- 50 1. Unadjusted Cost and Support
Fiscal costs of bank restructuring

tainment phase), step two—repair the damage (resolu-
tion phase). The principal forms of intervention were 40
y = 0.639x + 1.740
(percent of GDP)
(1) liquidity provision through collateralized lending R 2 = 0.223
30
and other arrangements; (2) support for short-term
wholesale funding markets; (3) (more extensive) 20
guarantees of retail deposits and other liabilities; (4)
purchases or exchanges of nonperforming or illiquid 10
assets; and (5) capital injections to banks. Interven-
tions often started with liquidity support to relieve the 0
0 10 20 30 40 50
immediate pressure and then moved on to identifying Liquidity support
and meeting recapitalization needs. (% of total deposits and liabilities to nonresidents)
Yet the timing and strength of the response varied
across countries, especially when it came to the chal- 2. Cost and Support Adjusted for Crisis
lenge of repairing the damage (Figure 2.3.1). Part of 7 Severity
Fiscal costs of bank restructuring

the variation certainly reflected when and how severely 6
a country was affected, plus how large the banking sec-
(divided by peak NPL)
5
tor was relative to GDP, but there are differences even
after controlling for crisis severity. Specific forms of 4
y = 0.247x + 0.808
intervention also differed. Some governments acquired 3 R 2 = 0.059
minority stakes in distressed banks while others chose 2
to close or nationalize them. Stress tests were intro-
1
duced to restore confidence, with different approaches
in design and governance. Sometimes, but not always, 0
0 1 2 3 4 5 6 7
measures aiming to reduce debt overhang in the Liquidity support
nonfinancial sector accompanied the interventions (Divided by peak NPL)
targeted at the financial institutions. Last but not least,
cross-country differences in structural features, such as Sources: Laeven and Valencia (2013); and IMF staff
resolution frameworks, bankruptcy regimes, and the calculations.
Note: To adjust for crisis severity, fiscal costs of bank
degree to which the system depended on bank- versus restructuring and liquidity support are divided by the peak
market-based financing, came into focus. NPL. NPL = nonperforming loan ratio.
Drawing on this variation, several insights can be
gained from comparing crisis management in the
United States and in Europe: • The resolution phases diverged more, laying bare
• The containment phases were fairly similar. The stark differences in regulatory and supervisory archi-
major central banks were quick to offer liquidity tecture across the two regions.
support through traditional facilities and established • The United States mobilized recapitalization plans
unconventional facilities to ensure that pressure in faster than did countries in the European Union
funding markets subsided. They also established (EU) (Figure 2.3.2).
swap lines as early as December 2007 and extended • In addition to speed, the actions taken in the
these to other central banks as the crisis spread. In United States were more decisive. Banks replenished
many respects, the response in the containment their eroded capital base by issuing new equity
phase was better coordinated internationally during early in the crisis, whereas, in the EU, there was
the recent crisis than in past crises (Laeven and no matching effort (Figure 2.3.3). At least in part,
Valencia 2013). this was driven by the supervisory approach: US
banks were compelled to raise fresh capital (and
The author of this box is Deniz Igan. were able to do so because of support from the

Box 2.3 (continued)

Figure 2.3.2. Timing of Recapitalization Figure 2.3.3. New Share Issuance by Banks
(Months) (Percent)
5 1.2 United States European Union
4
1.0
3
2 0.8
1
0.6
0
–1 0.4
–2
0.2
–3
–4 0.0
IRL AUT BEL NLD GBR 2005 06 07 08 09 10 11 12 13
LVA DNK DEU LUX USA
Sources: Homar and van Wijnbergen (2015); and IMF staff
Sources: Laeven and Valencia (2013); and IMF staff calculations.
calculations. Note: New share issuance by banks is measured by the
Note: Timing is measured by the months between moments volume in percent of the consolidated balance sheet.
when liquidity support became extensive and implementation
of recapitalization. Data labels use International Organization
for Standardization (ISO) country codes.
released in the 2010 exercise. The scenarios were
criticized for being too benign and not capturing
the risk of sovereign default—a major concern
Federal Reserve and other agencies); EU banks at the time (Abramovich 2011).1 Moreover, the
were instructed to improve their risk-weighted newly created European Financial Stability Facility
capital ratios, but options were left open on how to (EFSF)—tasked with potential capital assistance—
do that. Faced with tight funding conditions and could offer funding to member states by selling
broader uncertainty, banks chose to cut lending and bonds rather than investing directly in banks.2
increase their sovereign debt holdings—which carry Finally, despite the seal of approval gained by pass-
a zero risk weight under Basel III. ing the stress tests, many banks continued to strug-
• Further, while stress tests were conducted on both gle. Taken together, these led markets to label the
sides of the Atlantic, market perceptions of what exercise a “nonevent” with no useful information
they accomplished differed. In the United States, content (Shah 2010).3 The EU experience under-
the Supervisory Capital Assessment Program aimed
to address uncertainty about the solvency of sys-
1Regulators reportedly chose not to include a default scenario
temic institutions (Bernanke 2009). Moreover, the
“partly because they said that a sovereign default was unlikely
Treasury Department committed to making capital and partly due to worries that it would send the wrong political
available to eligible banks. Test results were publicly message” (Enrich 2010).
available on a bank-by-bank basis, providing the 2The EFSF was succeeded by the European Stability Mech-
needed information to nervous markets (Fernandes, anism, which, under some conditions, can provide funding
Igan, and Pinheiro 2015). In the European Union, directly to recapitalize banks.
3Regulators will prefer to fully reveal banks’ capital short-
the Committee of European Banking Supervisors fall at times of crisis if they are able to recapitalize them, but
conducted two rounds of tests. Individual results will hold onto some information if they cannot recapitalize
were kept confidential in the 2009 round, though (Spargoli 2012).

Box 2.3 (continued)

scored the importance of credibility—established Figure 2.3.4. Nonperforming Loan Ratio
through independent governance, the requisite tech- (Percent)
nical expertise, and clearly communicated plans for
30 United States Ireland Spain
any backstop needs (Ong and Pazarbasioglu 2013).
• Because the epicenter of the crisis in many countries
was housing markets, mortgage defaults became 25
endemic. In the United States, the Making Home
Affordable (MHA) program was introduced in
20
2009 to help struggling homeowners (Chapter 3
of the April 2012 World Economic Outlook). The
refinancing program under the MHA program, 15
in particular, provided substantial welfare gains to
highly indebted households (Mitman 2016) and
10
boosted consumption (Agarwal and others 2015).
In European countries caught up in their own
credit-fueled housing boom-bust, there were no 5
corresponding widespread programs at the outset
of the crisis. Nonperforming loan ratios increased
0
more than in the United States and remain high 2005 06 07 08 09 10 11 12 13 14 15
(Figure 2.3.4).4
• More generally, many European countries continue Source: World Bank, Global Financial Development Database.
to grapple with large stocks of impaired assets a
decade after the onset of the crisis. A large conflu-
ence of factors—the global financial crisis hit many
hard and particularly hurt those with their own the concentration of debt overhang in small and
homegrown bubbles (Claessens and others 2010)— medium-sized enterprises, this further amplified the
exposed the monetary union’s incomplete architec- impact of the banks’ problems and debt overhang
ture and triggered a sovereign debt crisis, subjecting on investment and consumption. Inadequate capital
banks to a second round of shocks. The deep and buffers, prudential problems with collateral valuation
prolonged economic downturn that followed further and treatment of nonperforming loans, legal obstacles
weakened borrowers’ debt service capacity, leading to debt enforcement, loan restructuring and foreclo-
to an increase in loan defaults and large corporate sure, and a lack of distressed debt markets have been
and household debt overhangs. The nonperform- identified as primary obstacles to nonperforming
ing loans are concentrated most notably in small loan resolution (Aiyar and others 2015).
and medium-sized enterprises, which contribute • A related point of comparison between the US and
almost two-thirds of Europe’s output and employ- EU experiences involves the resolution framework for
ment and tend to rely more on bank financing than banks themselves. In the former, having an estab-
large firms. In addition, many European countries lished resolution authority that can act independently
have bank-based financial systems.5 Together with on the best option to resolve distressed banks (across
state borders)—the Federal Deposit Insurance
Corporation—helped ensure swift resolution of fail-
4Ireland and Spain were chosen for illustrative purposes as
ing banks (although dealing with systemic financial
they both had housing booms and busts and significant banking
distress. Other EU countries that could be used for direct
institutions required further action). In the latter, the
comparison (for example, Greece, Italy, Portugal) either did not troubles of the banking system started a search for
have a similar precrisis boom-bust pattern in housing markets, or new mechanisms that culminated in the creation of a
their experience was dominated by the sovereign debt crisis that single supervisor and a unified resolution framework
followed the global financial crisis. (Goyal and others 2013).
5Market-based economies experience significantly and durably
stronger rebounds than those that are bank-based; in particular,

The postcrisis paths for credit, investment, consump-
the more bank-based economies of continental Europe (Allard tion, and growth differed accordingly (Figure 2.3.5).
and Blavy 2011). The United States recovered faster and more strongly.

Box 2.3 (continued)

Figure 2.3.5. Postcrisis Paths The deleveraging phase, notably, has been shorter and
(Percent; 2007 = 100) more shallow—consistent with the importance of
repairing bank balance sheets in restoring growth.6
United States Ireland Spain Summing up, comparison of the US and European
experiences and cross-country studies highlights the
120 1. Bank Credit 2. Private 120 following:
Investment
• Swift and decisive action: Recapitalizing or resolving
100
100 banks shortly after the containment phase is key.
80 The alternative leads to zombification, with signifi-
80 cant macroeconomic costs. From a structural point
60 of view, resolution frameworks should aim to ensure
that such swift and decisive action is possible.
60
40
• Appropriate backstops: In extreme circumstances,
establishing credibility and preventing panic and
20 40
2000 04 08 12 16 2000 05 09 13 17 contagion may require use of public funds. In this
context, having enough fiscal room and mitigat-
140 3. Private 4. GDP 120 ing the sovereign-bank nexus become crucial. Any
Consumption actual use of these backstops, however, should be
120 a last-resort measure accompanied by appropriate
100 burden sharing and clear exit strategies to mini-
mize moral hazard, as well as the potential costs
100
associated with direct government involvement in
80 financial markets (for example, efficiency concerns).
80
60 60
2000 05 09 13 17 2000 05 09 13 17
6Other evidence corroborates this insight: early and decisive
Sources: Organisation for Economic Co-operation and recapitalization of distressed banks helps corporate investment
Development; World Bank, World Development Indicators recover (Sun and Tong 2015) and can take several years off the
database; and IMF staff calculations. duration of a recession (Homar and van Wijnbergen 2015).

References Allard, Julien, and Rodolphe Blavy. 2011. “Market Phoenixes

and Banking Ducks: Are Recoveries Faster in Market-Based
Abiad, Abdul, Ravi Balakrishnan, Petya Koeva Brooks, Daniel
Economies?” IMF Working Paper 11/213, International
Leigh, and Irina Tytell. 2009. “What’s the Damage? Medium-
Monetary Fund, Washington, DC.
Term Output Dynamics after Banking Crises.” IMF Working
Arai, Mahmood, and Roger Vilhelmsson. 2004. “Unemploy-
Paper 09/245, International Monetary Fund, Washington, DC.
ment-Risk Differentials between Immigrant and Native
Abramovich, Alexander. 2011. “Comparative Analysis of Stress
Workers in Sweden.” Industrial Relations 43: 690–98.
Testing in the United States and Europe.” North Carolina
Arnold, Nathaniel, Bergljot Barkbu, Elif Ture, Hou Wang, and
Banking Institute 15 (1), Article 16, University of North
Jiaxiong Yao. 2018. “A Central Fiscal Stabilization Capacity
Carolina School of Law, Chapel Hill, NC. for the Euro Area.” IMF Staff Discussion Note 18/03, Inter-
Acemoglu, Daron, and Autor, David. 2011. “Skills, Tasks and national Monetary Fund, Washington, DC.
Technologies: Implications for Employment and Earnings.” Aslam, Aqib, Patrick Blagrave, Eugenio Cerutti, Sung Eun Jung,
Handbook of Labor Economics, Elsevier. and Carolina Osorio-Buitron. Forthcoming. “Recessions and
Acemoglu, Daron, and David Autor. 2011. “Skills, Tasks, and Recoveries: Are EMs Different from AEs?” IMF Working
Technologies: Implications for Employment and Earnings.” In Paper, International Monetary Fund, Washington, DC.
Handbook of Labor Economics (4), edited by Orley Ashenfelter Auerbach, Alan. 2017. “Fiscal Policy.” Peterson Institute for
and David E. Card. Amsterdam: Elsevier. International Economics, Washington, DC.
Acemoglu, Daron, and Pascual Restrepo. 2017. “Robots and Jobs: Autor, David, and David Dorn. 2013. “The Growth of Low-Skill
Evidence from US Labor Markets.” NBER Working Paper Service Jobs and the Polarization of the U.S. Labor Market.”
23285, National Bureau of Economic Research, American Economic Review 103 (5): 1553–1597.
Cambridge, MA. Autor, David, Frank Levy, and Richard Murnane. 2003. “The
———. 2018. “Artificial Intelligence, Automation, and Work.” Skill Content of Recent Technological Change: An Empir-
NBER Working Paper 24196, National Bureau of Economic ical Exploration.” Quarterly Journal of Economics 118 (4):
Research, Cambridge, MA. 1279–333.
———. Forthcoming. “The Race between Man and Machine: Ball, Laurence. 2014. “Long-Term Damage from the Great
Implications of Technology for Growth, Factor Shares, and Recession in OECD Countries.” NBER Working Paper
Employment.” American Economic Review. 20185, National Bureau of Economic Research, Cambridge,
Adams, Zoe, Louise Bishop, and Simon Deakin. 2016. CBR MA.
Labour Regulation Index. Cambridge: Centre for Business Barnichon, Regis, Christian Matthes, and Alexander Ziegenbein.
Research. 2018. “The Financial Crisis at 10: Will We Ever Recover?”
Adler, Gustavo, Romain Duval, Davide Furceri, Sinem Kiliç Federal Reserve Bank of San Francisco Economic Letter
Çelik, Ksenia Koloskova, and Marcos Poplawski-Ribeiro. 2018–19, August.
2017. “Gone with the Headwinds: Global Productivity.” IMF Barth, James R., Gerard Caprio, Jr., and Ross Levine. 2013.
Staff Discussion Note 17/04, International Monetary Fund, “Bank Regulation and Supervision in 180 Countries from
Washington, DC. 1999 to 2011.” NBER Working Paper 18733, National
Agarwal, Sumit, Gene Amromin, Souphala Chomsisengphet, Bureau of Economic Research, Cambridge, MA.
Tomasz Piskorski, Amit Seru, and Vincent Yao. 2015. “Mort- Bayoumi, Tamim. 2017. Unfinished Business: The Unexplored
gage Refinancing, Consumer Spending, and Competition: Causes of the Financial Crisis and the Lessons Yet to Be Learned.
Evidence from the Home Affordable Refinancing Program.” New Haven, CT: Yale University Press.
NBER Working Paper 21512, National Bureau of Economic Berg, Andrew, Ed Buffie, and Felipe Zanna. 2017. “Should We
Research, Cambridge, MA. Fear the Robot Revolution? (The Correct Answer Is Yes).”
Ahuja, Ashvin, and Malhar S. Nabar. 2012. “Investment-Led Manuscript prepared for the Carnegie-Rochester NYU Con-
Growth in China: Global Spillovers.” IMF Working Paper ference Series.
12/267, International Monetary Fund, Washington, DC. Berger, Helge, Giovanni Dell’Ariccia, and Maurice Obstfeld.
Aiyar, Shekhar, Wolfgang Bergthaler, Jose M. Garrido, Anna Ily- 2018. “Revisiting the Economic Case for Fiscal Union in
ina, Andreas Jobst, Kenneth Kang, Dmitriy Kovtun, Yan Liu, the Euro Area.” Departmental Paper, Research Department,
Dermot Monaghan, and Marina Moretti. 2015. “A Strategy International Monetary Fund, Washington, DC.
for Resolving Europe’s Problem Loans.” IMF Staff Discussion Berger, Helge, Thomas Dowling, Sergi Lanau, Weicheng Lian,
Note 15/19, International Monetary Fund, Washington, DC. Mico Mrkaic, Pau Rabanal, and Marzie Taheri Sanjani. 2015.
Allard, Céline, Petya Koeva Brooks, John C. Bluedorn, Fabian “Steady as She Goes—Estimating Potential Output during
Bornhorst, Katharine Christopherson, Franziska Ohnsorge, Financial ‘Booms and Busts’.” IMF Working Paper 15/233,
Tigran Poghosyan, and an IMF Staff Team. 2013. “Toward a International Monetary Fund, Washington, DC.
Fiscal Union for the Euro Area.” IMF Staff Discussion Note Berkmen, S. Pelin, Gaston Gelos, Robert Rennhack, and James
13/09, International Monetary Fund, Washington, DC. P. Walsh. 2012. “The Global Financial Crisis: Explaining

Cross-Country Differences in the Output Impact.” Journal of ture and Fertility.” The ANNALS of the American Academy of
International Money and Finance 31: 42–59. Political and Social Science 650: 214–31.
Bernanke, Ben S. 2009. “Statement Regarding the Supervisory Cap- Chiacchio, Francesco, Georgios Petropoulos, and David Pichler.
ital Assessment Program.” May 7. http://www.federalreserve.gov/ 2018. “The Impact of Industrial Robots on EU Employment
newsevents/press/bcreg/bernankescap20090507.htm. and Wages: A Local Labour Market Approach.” Bruegel
———. 2015. The Courage to Act: A Memoir of a Crisis and Its Working Paper 02, Bruegel, Brussels.
Aftermath. New York, NY: W. W. Norton & Company, Inc. Claessens, Stijn, Giovanni Dell’Ariccia, Deniz Igan, and Luc
———. 2017. “Monetary Policy for a New Era.” Prepared for Laeven. 2010. “Cross-Country Experiences and Policy Impli-
conference titled “Rethinking Macroeconomic Policy,” cations from the Global Financial Crisis.” Economic Policy
Peterson Institute for International Economics, Washington April: 267–93.
DC, October 12–13. Clark, Ximena, Timothy J. Hatton, and Jeffrey G. Williamson.
———. 2018. “The Real Effects of Disrupted Credit: Evidence 2007. “Explaining U.S. Immigration, 1971–98.” Review of
from the Global Financial Crisis.” The Per Jacobsson Foun- Economics and Statistics 89 (2): 359–73.
dation Lecture, Bank for International Settlements, Basel, Congressional Budget Office (CBO). 2014. “Revisions to CBO’s
and forthcoming in Brookings Papers on Economic Activity, Projection of Potential Output since 2007.” https://www.cbo.
Brookings Institution, Washington, DC. gov/publication/45150.
Blanchard, Olivier, Eugenio Cerutti, and Lawrence Summers. Elsby, Michael W.L., Bart Hobijn, and Aysegul Sahin. 2013.
2015. “Inflation and Activity—Two Explorations and Their “Unemployment Dynamics in the OECD.” Review of Eco-
Monetary Policy Implications.” NBER Working Paper 21726, nomics and Statistics 95 (2): 530–48.
National Bureau for Economic Research, Cambridge, MA. Enrich, David. 2010. “EU Banks Survive Stress Test.” The Wall
Blanchard, Olivier, Giovanni Dell’Ariccia, and Paolo Mauro. Street Journal, July 24. http://online.wsj.com/article/SB10001
2010. “Rethinking Macroeconomic Policy.” IMF Staff Position 424052748703294904575384940544522582.html.
Note 10/03, International Monetary Fund, Washington, DC. Espenshade, Thomas J. 1994. “Can Immigration Slow US
Blanchard, Olivier, Hamid Faruqee, and Mitali Das. 2010. “The Population Aging?” Journal of Policy Analysis and Management
Initial Impact of the Crisis on Emerging Market Countries.” 13 (4): 759–68.
Brooking Papers on Economic Activity (Spring): 263–23. Fernald, John G. 2015. “Productivity and Potential Output
Blanchard, Olivier, and Daniel Leigh. 2013. “Growth Forecast before, during, and after the Great Recession.” NBER Macroe-
Errors and Fiscal Multipliers.” American Economic Review 103 conomics Annual 2014 29 (1): 1–51.
(3): 117–20. Fernandes, Marcelo, Deniz Igan, and Marcelo Pinheiro. 2015.
Blanchard, Olivier, and Lawrence H. Summers. 2017. “Rethink- “March Madness in Wall Street: (What) Does the Market
ing Stabilization Policy: Back to the Future.” Conference titled Learn from Stress Tests?” IMF Working Paper 15/271, Inter-
“Rethinking Macroeconomic Policy,” Peterson Institute for national Monetary Fund, Washington, DC.
International Economics, Washington, DC, October 12–13. Frey, Carl Benedikt, and Michael A. Osborne. 2017. “The Future of
Blinder, Alan S. 2013. After the Music Stopped: The Financial Employment: How Susceptible Are Jobs to Computerisation?”
Crisis, the Response, and the Work Ahead. New York, NY: The Technological Forecasting and Social Change 114 (C): 254–80.
Penguin Press. Furman, Jason. 2018. “The Fiscal Response to the Great
Borjas, George J. 1987. “Self-Selection and the Earnings of Recession: Steps Taken, Paths Rejected, and Lessons for Next
Immigrants.” American Economic Review 77 (4): 531–53. Time.” Brookings Institution.
Candelon, Bertrand, Alina Carare, Jean-Baptiste Hasse, and Jing Gagnon, Joseph E. 2016. “Quantitative Easing: An Underappre-
Lu. 2018. “Globalization and the New Normal.” IMF Work- ciated Success.” Peterson Institute of International Economics
ing Paper 18/75, International Monetary Fund, Washington, Policy Brief 16–4, Peterson Institute of International Eco-
DC. nomics, Washington, DC.
Cerra, Valerie, Ugo Panizza, and Sweta Saxena. 2013. “Interna- Geithner, Timothy F. 2014. Stress Test: Reflections on Financial
tional Evidence on Recovery from Recessions.” Contemporary Crises. New York, NY: Broadway Books.
Economic Policy 31 (2): 424–39. Gertler, Mark, and Simon Gilchrist. 2018. “What Happened:
Cerra, Valerie, and Sweta Saxena. 2008. “Growth Dynamics: The Financial Factors in the Great Recession.” NBER Working
Myth of Economic Recovery.” American Economic Review 98 Paper 24746, National Bureau of Economic Research, Cam-
(1): 439–57. bridge, MA.
———. 2017. “Booms, Crises, and Recoveries: A New Par- Giannone, Domenico, Michele Lenza, and Lucrezia Reichlin.
adigm of the Business Cycle and Its Policy Implications.” 2011. “Market Freedom and the Global Recession.” IMF
IMF Working Paper 17/250, International Monetary Fund, Economic Review 59 (1): 111–35.
Washington, DC. Goos, Maarten, and Alan Manning. 2007. “Lousy and Lovely
Cherlin, Andrew, Erin Cumberworth, and Christopher Morgan. Jobs: The Rising Polarization of Work in Britain.” The Review
2013. “The Effects of the Great Recession on Family Struc- of Economics and Statistics 89 (1): 118–133.

———, and Anna Salomons. 2014. “Explaining Job Polariza- Kalwij, Adriaan. 2010. “The Impact of Family Policy Expenditure
tion: Routine-Biased Technological Change and Offshoring.” on Fertility in Western Europe.” Demography 47 (2): 503–519.
American Economic Review 104 (8): 2509–26. Kozlowski, Julian, Laura Veldkamp, and Venky Venkateswaran.
Gourinchas, Pierre-Olivier, and Maurice Obstfeld. 2012. “Stories 2017. “The Tail That Wags the Economy: Beliefs and Per-
of the Twentieth Century for the Twenty-First.” American sistent Stagnation.” NBER Working Paper 21719, National
Economic Journal: Macroeconomics 4 (1): 226–65. Bureau of Economic Research. Cambridge, MA.
Goyal, Rishi, Petya Koeva Brooks, Mahmood Pradhan, Thierry Laeven, Luc, and Fabián Valencia. 2013. “Systemic Banking
Tressel, Giovanni Dell’Ariccia, Ross Leckow, Ceyla Pazarba- Crises Database.” IMF Economic Review 61 (2): 225–70.
sioglu, and an IMF Staff Team. 2013. “A Banking Union for Lagarde, Christine. 2014. “The Challenge Facing the Global
the Euro Area.” IMF Staff Discussion Note 13/01, Interna- Economy: New Momentum to Overcome a New Mediocre.”
tional Monetary Fund, Washington, DC. Speech at Georgetown University School of Foreign Service,
Graetz, Georg, and Guy Michaels. Forthcoming. “Robots at Washington, DC, October 2.
Work.” Review of Economics and Statistics. ———. 2016. “Decisive Action to Secure Durable Growth.”
Hall, Robert E. 2014. “Quantifying the Lasting Harm to the US Lecture at an event hosted by Bundesbank and Goethe Uni-
Economy from the Financial Crisis.” NBER Macroeconomics versity, Frankfurt, Germany, April 5.
Annual 29: 71–128. Lane, Philip R., and Gian Maria Milesi-Ferretti. 2010. “The
Hatton, Timothy J., and Jeffrey G. Williamson. 2002. “What Cross-Country Incidence of the Global Crisis.” IMF Work-
Fundamentals Drive World Migration?” NBER Working Paper ing Paper 10/171, International Monetary Fund, Washing-
9159, National Bureau of Economic Research, ton, DC.
Cambridge, MA. ———. 2014. “Global Imbalances and External Adjustment
Hoem, Jan M. 2008. “The Impact of Public Policies on Euro- after the Crisis.” IMF Working Paper 14/151, International
pean Fertility.” Demographic Research (19): 249–60. Monetary Fund, Washington, DC.
Homar, Timotej, and Sweder J. G. van Wijnbergen. 2015. “On ———. 2017. “International Financial Integration in the After-
Zombie Banks and Recessions after Systemic Banking Crises: math of the Global Financial Crisis.” IMF Working Paper
Government Intervention Matters.” CEPR Discussion Paper 17/115, International Monetary Fund, Washington, DC.
DP10963, Centre for Economic Policy Research, London. Lee, Jaewoo, Jonathan D. Ostry, Alessandro Prati, Luca A. Ricci,
International Federation of Robotics (IFR). 2017. “World and Gian Maria Milesi-Ferretti. 2008. “Exchange Rate Assess-
Robotics 2017 Industrial Robots.” https://ifr.org. ments: CGER Methodologies, IMF Occasional Paper 261,
International Monetary Fund (IMF). 2009. “The State of International Monetary Fund.” Washington, DC.
Public Finances: Outlook and Medium-Term Policies after Lewis, Michael. 2010. The Big Short: Inside the Doomsday
the 2008 Crisis.” Fiscal Affairs Department publication, Machine. New York, NY: W. W. Norton & Company, Inc.
Washington, DC. Lipton, David. 2018. “Trust and the Future of Multilateralism.”
———. 2012. “Euro Area Policies: 2012 Article IV Consulta- IMF Blog.
tion.” IMF Country Report 12/181, Washington, DC. Lowenstein, Roger. 2010. The End of Wall Street. New York, NY:
———. 2013a. “Euro Area Policies: 2013 Article IV Consulta- The Penguin Press.
tion—Staff Report.” IMF Country Report 13/231, Washing- Mantega, Guido. 2010. International Monetary and Financial
ton, DC. Committee, Statement by Guido Mantega, Minister of
———. 2013b. “Greece: Ex Post Evaluation of Exceptional Finance, Ministerio da Fazenda, Brazil, October 9.
Access under the 2010 Stand-By Arrangement.” IMF Coun- Mitman, Kurt. 2016. “Macroeconomic Effects of Bankruptcy
try Report 13/156, Washington, DC. and Foreclosure Policies.” American Economic Review 106 (8):
———. 2013c. “Reassessing the Role and Modalities of Fiscal 2219–55.
Policy in Advanced Economies.” IMF Policy Paper, Washing- Neels, Karel. 2010. “Temporal Variation in Unemployment
ton, DC. Rates and Their Association with Tempo and Quantum of
———. 2014. “IMF Multilateral Policy Issues Report.” IMF Fertility: Some Evidence for Belgium, France, and the Neth-
Spillover Report, Washington, DC. erlands.” Paper prepared for the session titled “Low Fertility
———. 2015. “Crisis Program Review.” Strategy, Policy and and Its Association with Macroeconomic Trends” of the
Review Department, International Monetary Fund, Washing- Annual Meeting of the Population Association of America.
ton, DC. Obstfeld, Maurice, and Kenneth Rogoff. 2009. “Global Imbal-
Jordà, Òscar, Moritz Schularick, and Alan M. Taylor. 2016. ances and the Financial Crisis: Products of Common Causes.”
“Sovereigns versus Banks: Credit, Crises, and Consequences.” Proceedings, Federal Reserve Bank of San Francisco, October
Journal of the European Economic Association 14 (1): 45–79. 2009.
Kahanec, Martin, and Martin Guzi. 2017. “How Immigrants Ong, Li Lian, and Ceyla Pazarbasioglu. 2013. “Credibility and
Helped EU Labor Markets to Adjust during the Great Reces- Crisis Stress Testing.” IMF Working Paper 13/178, Interna-
sion.” International Journal of Manpower 38 (7): 996–1015. tional Monetary Fund, Washington, DC.

Organisation for Economic Co-operation and Development Sorkin, Andrew Ross. 2009. Too Big to Fail: The Inside Story of
(OECD). 2009. International Migration Outlook. How Wall Street and Washington Fought to Save the Financial
Paulson, Jr., Henry H. 2013. On the Brink: Inside the Race to System—and Themselves. New York, NY: Viking Penguin.
Stop the Collapse of the Global Financial System. New York, Spargoli, Fabrizio. 2012. “Bank Recapitalization and the
NY: Business Plus. Information Value of a Stress Test in a Crisis.” Manuscript,
Perri, Fabrizio, and Vincenzo Quadrini. 2018. “International Universitat Pompeu Fabra.
Recessions.” American Economic Review 108 (4–5): 935–84. Summers, Lawrence H. 2016. “The Age of Secular Stagnation:
Pritchett, Lant, and Lawrence H. Summers. 2013. “Asia-Phoria What It Is and What to Do about It.” Foreign Affairs, March/
Meet Regression to the Mean.” Proceedings, Federal Reserve April.
Bank of San Francisco (November): 1–35. Sun, Yangfang, and Hui Tong. 2015. “How Does Postcrisis
Rajan, Raghuram G. 2010. Fault Lines: How Hidden Fractures Bank Capital Adequacy Affect Firm Investment?” IMF
Still Threaten the Global Economy. Princeton, NJ: Princeton Working Paper 15/145, International Monetary Fund,
University Press. Washington, DC.
———. 2014. “Competitive Monetary Easing: Is It Yesterday Thévenon, Olivier. 2011. “Family Policies in OECD Countries:
Once More?” Remarks at the Brookings Institution, Washing- A Comparative Analysis.” Population and Development Review
ton, DC, April 10. 37 (1): 57–87.
Romer, Christina D., and David H. Romer. 2018. “Phillips Thomsen, Poul. 2017. “The Euro Zone: What’s Next.” Speech
Lecture—Why Some Times Are Different: Macroeconomic at the Financial Times Investment Management Summit,
Policy and the Aftermath of Financial Crises.” Economica 85: London, September 28.
1–40. Timmer, Marcel. P., Erik Dietzenbacher, Bart Los, Robert
Schanzenbach, Diane, Ryan Nunn, Lauren Bauer, and Audrey Stehrer, and Gaaitzen J. de Vries. 2015. “An Illustrated User
Breitwieser. 2017. “The Closing of the Jobs Gap: A Decade Guide to the World Input-Output Database: The Case of
of Recession Recovery.” Brookings Institution, The Hamilton Global Automotive Production.” Review of International
Project, Washington, DC. Economics 23 (3): 575–605.
Schwartz, Carl, and Nicholas Tan. 2016. “The Australian Toloui, Ramin. 2018. “Global Financial Crisis: Origins and
Government Guarantee Scheme: 2008–15.” Bulletin – March Causes.” Lectures Series on Navigating Financial Crises
Quarter, Reserve Bank of Australia. in the Modern Global Economy, Stanford University,
Shah, Neil. 2010. “Wait-and-See after Europe Stress Test.” Stanford, CA.
The Wall Street Journal, July 26. http://blogs.wsj.com/ Tsangarides, Charalambos. 2012. “Crisis and Recovery: Role of
marketbeat/2010/07/26/wait-and-see-aftereurope-stress-test/. the Exchange Rate Regime in Emerging Market Economies.”
Sobotka, Tomáš, Vegard Skirbekk, and Dimiter Philipov. 2011. Journal of Macroeconomics 34: 470–88.
“Economic Recession and Fertility in the Developed World.” Zhou, Xiaochuan. 2010. International Monetary and Financial
Population and Development Review 37 (2): 267–306. Committee, Statement by Zhou Xiaochuan, Governor of the
Solt, Frederick. 2016. “The Standardized World Income Inequal- People’s Bank of China and Governor of the IMF for China,
ity Database.” Social Science Quarterly 97 (5): 1267–281. October 9.

The Global Economic Recovery 10 Years After The 2008 Financial Meltdown

Uploaded by

Copyright:

Available Formats

The Global Economic Recovery 10 Years After The 2008 Financial Meltdown

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Global Economic Recovery 10 Years After The 2008 Financial Meltdown

Uploaded by

Copyright:

Available Formats

2

International Monetary Fund | October 2018 71

72 International Monetary Fund | October 2018

Following the global financial meltdown in late

ences when estimating an underlying trend path for

International Monetary Fund | October 2018 73

6 Banking crisis No banking crisis

74 International Monetary Fund | October 2018

9An important exception is China, where the investment share

of GDP rose from below 40 percent in precrisis years to almost

International Monetary Fund | October 2018 75

Two important consequences of sluggish invest-

76 International Monetary Fund | October 2018

Slower Technology Adoption

International Monetary Fund | October 2018 77

Sources: International Federation of Robotics; World Input-Output Database; and

13Analysis at the industry-country level (Online Annex 2.3.B)

78 International Monetary Fund | October 2018

tions of banking disruption and household balance sheets to the

International Monetary Fund | October 2018 79

80 International Monetary Fund | October 2018

International Monetary Fund | October 2018 81

82 International Monetary Fund | October 2018

2. Crisis-Related Discretionary Fiscal Stimulus in G20 Economies (as of October 2010)

International Monetary Fund | October 2018 83

Figure 2.12. Impact on 2015–17 GDP Deviations from Summary

ﬁnancial and other sectors

countries hit harder by the crisis.

84 International Monetary Fund | October 2018

International Monetary Fund | October 2018 85

Box 2.1. The Global Financial Crisis, Migration, and Fertility

The Great Recession and Migration –1

The decades leading up to the global finan- –2

periods, while low-income developing countries have 80

minus the number of emigrants over a period, divided by the

86 International Monetary Fund | October 2018

Box 2.1 (continued)

from source country over a period, divided by 1,000 population

International Monetary Fund | October 2018 87

Box 2.1 (continued)

fertility rates generally converge to that of natives

88 International Monetary Fund | October 2018

Box 2.1 (continued)

2. Effect of Policy Changes

Sources: Organisation for Economic Co-operation and

International Monetary Fund | October 2018 89

90 International Monetary Fund | October 2018

Box 2.2 (continued)

International Monetary Fund | October 2018 91

Box 2.2 (continued)

0.20 2. Low Output Losses

Sources: Cambridge University Center for Business Research

92 International Monetary Fund | October 2018

Fiscal costs of bank restructuring

Fiscal costs of bank restructuring

International Monetary Fund | October 2018 93

Box 2.3 (continued)

5 1.2 United States European Union

94 International Monetary Fund | October 2018