51 growth in time debt aer

8/2/2019 51 Growth in Time Debt Aer

1/6

573

American Economic Review: Papers & Proceedings 100 (May 2010): 573578

http://www.aeaweb.org/articles.php?doi=10.1257/aer.100.2.573

In this paper, we exploit a new multi-countryhistorical dataset on public (government) debt tosearch or a systemic relationship between highpublic debt levels, growth and infation.1 Ourmain result is that whereas the link betweengrowth and debt seems relatively weak at nor-mal debt levels, median growth rates or coun-tries with public debt over roughly 90 percento GDP are about one percent lower than other-

wise; average (mean) growth rates are severalpercent lower. Surprisingly, the relationshipbetween public debt and growth is remarkablysimilar across emerging markets and advancedeconomies. This is not the case or infation. Wend no systematic relationship between highdebt levels and infation or advanced econo-mies as a group (albeit with individual countryexceptions including the United States). By con-trast, in emerging market countries, high publicdebt levels coincide with higher infation.

Our topic would seem to be a timely one.Public debt has been soaring in the wake o therecent global nancial maelstrom, especially inthe epicenter countries. This should not be sur-prising, given the experience o earlier severenancial crises.2 Outsized decits and epic bankbailouts may be useul in ghting a downturn,but what is the long-run macroeconomic impact,

1 In this paper public debt reers to gross centralgovernment debt. Domestic public debt is governmentdebt issued under domestic legal jurisdiction. Public debtdoes not include debts carrying a government guarantee.Total gross external debt includes the external debts oallbranches o government as well as private debt that is issuedby domestic private entities under a oreign jurisdiction.

2 Reinhart and Rogo(2009a, b) demonstrate that theatermath o a deep nancial crisis typically involves aprotracted period o macroeconomic adjustment, particu-

larly in employment and housing prices. On average, publicdebt rose by more than 80 percent within three years atera crisis.

GrowthinaTimeofDebt

By C M. R K S. R*

especially against the backdrop o graying pop-ulations and rising social insurance costs? Aresharply elevated public debts ultimately a man-ageable policy challenge?

Our approach here is decidedly empirical,taking advantage o a broad new historicaldataset on public debt (in particular, centralgovernment debt) rst presented in Carmen M.Reinhart and Kenneth S. Rogo(2008, 2009b).

Prior to this dataset, it was exceedingly dicultto get more than two or three decades o pub-lic debt data even or many rich countries, andvirtually impossible or most emerging markets.Our results incorporate data on 44 countriesspanning about 200 years. Taken together, thedata incorporate over 3,700 annual observationscovering a wide range o political systems, insti-tutions, exchange rate and monetary arrange-ments, and historic circumstances.

We also employ more recent data on externaldebt, including debt owed both by governmentsand by private entities. For emerging markets,we nd that there exists a signicantly moresevere threshold or total gross external debt(public and private)which is almost exclu-sively denominated in a oreign currencythanor total public debt (the domestically issuedcomponent o which is largely denominatedin home currency). When gross external debtreaches 60 percent o GDP, annual growthdeclines by about two percent; or levels oexternal debt in excess o 90 percent o GDP,

growth rates are roughly cut in hal. We are notin a position to calculate separate total exter-nal debt thresholds (as opposed to public debtthresholds) or advanced countries. The avail-able time-series is too recent, beginning only in2000. We do note, however, that external debtlevels in advanced countries now average nearly200 percent o GDP, with external debt levelsbeing particularly high across Europe.

The ocus o this paper is on the longer termmacroeconomic implications o much higher

public and external debt. The nal section, how-ever, summarizes the historical experience othe United States in dealing with private sector

* Reinhart: Department o Economics, 4115 TydingsHall, University o Maryland, College Park, MD 20742

(e-mail: [email protected]); Rogo: Economics Depart-

ment, 216 Littauer Center, Harvard University, CambridgeMA 021383001 (e-mail: [email protected]). Theauthors would like to thank Olivier Jeanne and Vincent R.Reinhart or helpul comments.


2/6

MAY 2010574 AEA PAPERS AND PROCEEDINGS

deleveraging o debts, normal ater a nancialcrisis. Not surprisingly, such episodes are asso-ciated with very slow growth and defation.

I. The 20072009 Global Buildup in Public Debt

Figure 1 illustrates the increase in (infation-adjusted) public debt that has occurred since

2007. For the ve countries with systemic nan-cial crises (Iceland, Ireland, Spain, the UnitedKingdom, and the United States), average debtlevels are up by about 75 percent, well on track toreach or surpass the three year 86 percent bench-mark that Reinhart and Rogo (2009a,b), ndor earlier deep postwar nancial crises. Even incountries that did not experience a major nan-cial crisis, debt rose by an average o about 20percent in real terms between 2007 and 2009.3

3 Our ocus on gross central government debt owes tothe act that time series o broader measures o government

This general rise in public indebtedness stands instark contrast to the 20032006 period o pub-lic deleveraging in many countries and owes todirect bailout costs in some countries, the adop-tion o stimulus packages to deal with the globalrecession in many countries, and marked declinesin government revenues that have hit advancedand emerging market economies alike.

II. Debt,Growth,andInfation

The nonlinear eect o debt on growth isreminiscent o debt intolerance (Reinhart,Rogo, and Miguel A. Savastano 2003) andpresumably is related to a nonlinear responseo market interest rates as countries reach debt

tolerance limits. Sharply rising interest rates,in turn, orce painul scal adjustment in theorm o tax hikes and spending cuts, or, insome cases, outright deault. As or infation,an obvious connection stems rom the act thatunanticipated high infation can reduce thereal cost o servicing the debt. O course, theecacy o the infation channel is quite sen-sitive to the maturity structure o the debt. Inprinciple, the manner in which debt builds upcan be important. For example, war debts arearguably less problematic or uture growthand infation than large debts that are accu-mulated in peacetime. Postwar growth tendsto be high as wartime allocation o manpowerand resources unnels to the civilian economy.Moreover, high wartime government spending,typically the cause o the debt buildup, comesto a natural close as peace returns. In contrast,a peacetime debt explosion oten refects unsta-ble political economy dynamics that can persistor very long periods.

Here we will not attempt to determine the gen-

esis o debt buildups but instead simply look attheir connection to average and median growthand infation outcomes. This may lead us, i any-thing, to understate the adverse growth implica-tions o debt burdens arising out o the currentcrisis, which was clearly a peacetime event.

debt are not available or many countries. O course, thetrue run-up in debt is signicantly larger than stated here,at least on a present value actuarial basis, due to the exten-

sive government guarantees that have been conerred on thenancial sector in the crisis countries and elsewhere, whereor example deposit guarantees were raised in 2008.

69

44

72

42

84

62

22

9

21

29

25

47

119

47

62

4

44

182

46

32

41

49

Debt/GDP

2009100 150 200 250

Iceland

Ireland

UK

Spain

US

Crisis country average

Norway

Australia

China

Thailand

Mexico

Malaysia

Greece

Canada

Austria

Chile

Germany

Japan

Brazil

Korea

India

Average for others

2007 = 100

175.1

(increase of 75%)

120 (increase of 20%)

Figure 1. Cumulative Increase in Real Public Debt

Since 2007, Selected Countries

Note: Unless otherwise noted these gures are or centralgovernment debt defated by consumer prices.

Sources: Prices and nominal GDP rom InternationalMonetary Fund, World Economic Outlook. For a completelisting o sources or government debt, see Reinhart and

Rogo(2009b).


3/6

VOL. 100 NO. 2 575GRO IN A IME Of DEB

A.EvidenceromAdvancedCountries

Figure 2 presents a summary o infation andGDP growth across varying levels o debt or 20advanced countries over the period 19462009.This group includes Australia, Austria, Belgium,Canada, Denmark, Finland, France, Germany,Greece, Ireland, Italy, Japan, Netherlands, NewZealand, Norway, Portugal, Spain, Sweden, theUnited Kingdom, and the United States. Theannual observations are grouped into our cat-egories, according to the ratio o debt to GDPduring that particular year as ollows: years when

debt to GDP levels were below 30 percent (lowdebt); years where debt/GDP was 30 to 60 per-cent (medium debt); 60 to 90 percent (high); and

above 90 percent (very high). The bars in Figure2 show average and median GDP growth oreach o the our debt categories. Note that o the1,186 annual observations, there are a signicantnumber in each category, including 96 above 90percent. (Recent observations in that top bracketcome rom Belgium, Greece, Italy, and Japan.)From the gure, it is evident that there is noobvious link between debt and growth until pub-lic debt reaches a threshold o 90 percent. Theobservations with debt to GDP over 90 percenthave median growth roughly 1 percent lower thanthe lower debt burden groups and mean levels ogrowth almost 4 percent lower. (Using laggeddebt does not dramatically change the picture.)The line in Figure 2 plots the median infation or

the dierent debt groupingswhich makes plainthat there is no apparent pattern osimultaneousrising infation and debt.

Table 1 provides detail on the growth experi-ence or individual countries, but over a muchlonger period, typically one to two centuries.Interestingly, introducing the longer time-seriesyields remarkably similar conclusions. Over thepast two centuries, debt in excess o 90 percenthas typically been associated with mean growtho 1.7 percent versus 3.7 percent when debt islow (under 30 percent o GDP), and comparedwith growth rates o over 3 percent or the twomiddle categories (debt between 30 and 90 per-cent o GDP). O course, there is considerablevariation across the countries, with some coun-tries such as Australia and New Zealand experi-encing no growth deterioration at very high debtlevels. It is noteworthy, however, that those high-growth high-debt observations are clustered inthe years ollowing World War II.

B.EvidenceromEmergingMarketCountries

We next perorm the same exercise or 24emerging market economies or the periods19462009 and 19002009, using comparablecentral government debt data to those we usedor the advanced economies.4 Perhaps surpris-ingly, the results illustrated in Figure 2 andTable 1 or advanced economies are repeatedor emerging market economies. The emerging

4

While we have pre-1900 infation, real GDP, and publicdebt data or many emerging markets, nominal GDP data isharder to nd.

1.0

0.0

1.0

2.0

3.0

4.0

5.0

Average Median Average Median Average Median Average Median

GDP

growth

2

2.5

3

3.5

4

4.5

5

5.5

6

Inflation

Debt/GDP

below 30%

Debt/GDP

30 to 60%

Debt/GDP

60 to 90%

Debt/GDP

above 90%

Inflation(line, right axis)

GDP growth (bars, left axis)

Figure 2. Government Debt, Growth, and Inflation:

Selected Advanced Economies, 19462009

Notes: Central government debt includes domestic andexternal public debts. The 20 advanced economies includedare Australia, Austria, Belgium, Canada, Denmark,Finland, France, Germany, Greece, Ireland, Italy, Japan,Netherlands, New Zealand, Norway, Portugal, Spain,Sweden, the United Kingdom, and the United States. The

number o observations or the our debt groups are: 443or debt/GDP below 30 percent; 442 or debt/GDP 30 to 60percent; 199 observations or debt/GDP 60 to 90 percent;and 96 or debt/GDP above 90 percent. There are 1,180observations.Sources: International Monetary Fund, orldEconomicOutlook, OECD, World Bank, Global Developmentfinance, and Reinhart and Rogo (2009b) and sourcescited therein.


4/6


market equivalents oFigure 2 and Table 1 arenot reproduced here (to economize on space),

but the interested reader is reerred to theNBER working paper version o this paper.For 19002009, or example, median and aver-age GDP growth hovers around 44.5 percentor levels o debt below 90 percent o GDP, butmedian growth alls markedly to 2.9 percentor high debt (above 90 percent); the decline iseven greater or the average growth rate, whichalls to 1 percent. With much aster populationgrowth than the advanced economies, the impli-cations or per capita GDP growth are in line (or

worse) with those shown or advanced econo-mies. The similarities with advanced economiesend there, as higher debt levels are associated

with signicantly higher levels o infation inemerging markets. Median infation more than

doubles (rom less than seven percent to 16 per-cent) as debt rises rom the low (0 to 30 percent)range to above 90 percent. Fiscal dominance is aplausible interpretation o this pattern.

Because emerging markets oten depend somuch on external borrowing, it is interesting tolook separately at thresholds or external debt(public and private). In Figure 3, we highlightthe connection between gross external debt asreported by the World Bank and growth andinfation. As one can see, the growth thresholds

or external debt are considerably lower than thethresholds or total public debt. Growth dete-riorates markedly at external debt levels over

Table 1Real GDP Growth as the Level of Government Debt Varies:

Selected Advanced Economies, 17902009

(annual percent change)

Central (ederal) government debt/GDP

Country Period Below 30percent 30 to 60percent 60 to 90percent 90 percentand above

Australia 19022009 3.1 4.1 2.3 4.6Austria 18802009 4.3 3.0 2.3 n.a.Belgium 18352009 3.0 2.6 2.1 3.3Canada 19252009 2.0 4.5 3.0 2.2Denmark 18802009 3.1 1.7 2.4 n.a.Finland 19132009 3.2 3.0 4.3 1.9France 18802009 4.9 2.7 2.8 2.3Germany 18802009 3.6 0.9 n.a. n.a.Greece 18842009 4.0 0.3 4.8 2.5Ireland 19492009 4.4 4.5 4.0 2.4Italy 18802009 5.4 4.9 1.9 0.7Japan 18852009 4.9 3.7 3.9 0.7

Netherlands 18802009 4.0 2.8 2.4 2.0New Zealand 19322009 2.5 2.9 3.9 3.6Norway 18802009 2.9 4.4 n.a. n.a.Portugal 18512009 4.8 2.5 1.4 n.a.Spain 18502009 1.6 3.3 1.3 2.2Sweden 18802009 2.9 2.9 2.7 n.a.United Kingdom 18302009 2.5 2.2 2.1 1.8United States 17902009 4.0 3.4 3.3 1.8

Average 3.7 3.0 3.4 1.7Median 3.9 3.1 2.8 1.9

Observations = 2,317 866 654 445 352

Notes: An n.a. denotes no observations were recorded or that particular debt range. Thereare missing observations, most notably during World War I and II years; urther details areprovided in the data appendices to Reinhart and Rogo (2009b) and are available rom theauthors. Minimum and maximum values or each debt range are shown inbolded i talics.Sources: There are many sources; among the more prominent are: International MonetaryFund, orld Economic Outlook, OECD, World Bank, Global Development finance. Extensiveother sources are cited in Reinhart and Rogo(2009).


5/6

VOL. 100 NO. 2 577 GRO IN A IME Of DEB

60 percent, and urther still when external debtlevels exceed 90 percent, which record outrightdeclines. In light o this, it is more understand-able that over one hal o all deaults on externaldebt in emerging markets since 1970 occurred at

levels o debt that would have met the Maastrichtcriteria o 60 percent. Infation becomes signi-cantly higher only or the group o observationswith external debt over 90 percent.

III. PrivateSectorDebt:AnIllustration

Our main ocus has been on central govern-ment debt and, to a lesser degree, external publicand private debt, since reliable data on privatedomestic debts are much scarcer across countries

and time. We have argued here and elsewherethat a key legacy o a deep nancial crisis israpidly expanding public debt. Furthermore, we

have shown that public levels o debt/GDP thatpush the 90 percent threshold are associated withlower median and average growth.5 These obser-vations, however, present only a partial picture othe post-nancial crisis landscape. Private debt,in contrast, tends to shrink sharply in the ater-math o a nancial crisis. Just as a rapid expan-sion in private credit uels the boom phase o thecycle, so does serious deleveraging exacerbate thepost-crisis downturn. This pattern is illustrated inFigure 4, which shows the ratio o private debt toGDP or the United States or 19162009. Periodso sharp deleveraging have ollowed periods olower growth and coincide with higher unem-ployment. In varying degrees, the private sector(households and rms) in many other countries

(notably both advanced and emerging Europe)are also unwinding the debt built up during theboom year. Thus, private deleveraging may beanother legacy o the nancial crisis that maydampen growth in the medium term.

IV. ConcludingRemarks

The sharp run-up in public sector debt willlikely prove one o the most enduring lega-cies o the 20072009 nancial crises in theUnited States and elsewhere. We examine theexperience o 44 countries spanning up to twocenturies o data on central government debt,infation and growth. Our main nding is thatacross both advanced countries and emergingmarkets, high debt/GDP levels (90 percent andabove) are associated with notably lower growthoutcomes. Much lower levels o external debt/GDP (60 percent) are associated with adverseoutcomes or emerging market growth. Seldomdo countries grow their way out o debts. Thenonlinear response o growth to debt as debt

grows towards historical boundaries is remi-niscent o the debt intolerance phenomenondeveloped in Reinhart, Rogo, and Savastano(2003). As countries hit debt tolerance ceilings,market interest rates can begin to rise quite sud-denly, orcing painul adjustment.

O course, there are other vulnerabilitiesassociated with debt buildups, particularly igovernments try to mitigate servicing costs by

5

It is important to note that post-crises increases in pub-lic debt do not necessarily push economies into the vulner-able 90+ debt/GDP range.

1.5

0.5

0.5

1.5

2.5

3.5

4.5

5.5

Average Median Average Median Average Median Average Median

GDP

growth

10

11

12

13

14

15

16

17

Debt/GDP

below 30%

Debt/GDP

30 to 60% Debt/GDP60 to 90% Debt/GDPabove 90%

Inflation(line, right axis)

GDP growth (bars, left axis)

Inflation

Figure 3. External Debt, Growth, and Inflation:

Selected Emerging Markets, 1970-2009

Notes: The 20 emerging market countries included areArgentina, Bolivia, Brazil, Chile, China, Colombia, Egypt,India, Indonesia, Korea, Malaysia, Mexico, Nigeria, Peru,Philippines, South Arica, Thailand, Turkey, Uruguay, andVenezuela. The number o observations or the our debtgroups are: 252 or debt/GDP below 30 percent; 309 ordebt/GDP 30 to 60 percent; 120 observations or debt/GDP60 to 90 percent; and 74 or debt/GDP above 90 percent.There is a total o 755 annual observations.Sources: International Monetary Fund, orldEconomicOutlook, World Bank, GlobalDevelopmentfinance, andReinhart and Rogo(2009b) and sources cited therein.


6/6


shortening the maturing structure o debt. AsReinhart and Rogo (2009b) emphasize andnumerous models suggest, countries that chooseto rely excessively on short-term borrowing tound growing debt levels are particularly vul-nerable to crises in condence that can provokevery sudden and unexpected nancial crises.At the very minimum, this would suggest thattraditional debt management issues should be atthe oreront o public policy concerns.

REFERENCES

Reinhart, Carmen M., and Kenneth S. Rogo.2008. The Forgotten History o Domestic

Debt. National Bureau o Economic ResearchWorking Paper 13946.

Reinhart, Carmen M., and Kenneth S. Rogo.

2009a. The Atermath o Financial Cri-ses. American Economic Review, 99(2):46672.

Reinhart, Carmen M., and Kenneth S. Rogo.

2009b. his ime Is Dierent: Eight Centurieso financial folly. Princeton, NJ: PrincetonUniversity Press.

Reinhart, Carmen M., Kenneth S. Rogo, and

Miguel A. Savastano. 2003. Debt Intoler-ance.Brookings Papers on Economic Activ-

ity (1), ed. William C. Brainard and George L.Perry, 162.

19161939 19462009

Years with debt/GDP declines 9.8 7.2All other years 6.7 5.5

unemployment rateMedian

20

70

120

170

220

270

320

1916 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006

Historical statistics

of the United States

Flow of funds

Percent

Figure 4. United States: Private Debt Outstanding, 19162009

(end-of-period stock of debt as a percent of GDP)

Note: Data or 2009 is end-o-June.

Sources:istorical Stat istics o the United States, Flow o Funds, Board o Governors o theFederal Reserve, International Monetary Fund, orld Economic Outlook, OECD, World Bank,Global Development finance, and Reinhart and Rogo(2009b) and sources cited therein.

51 growth in time debt aer

Documents