Sovereign default which shocks matter

Of these elements, the two most innovative conceptual contributions are the introduction of the return differential motive for borrowing into the sovereign default literature and the explanation for why default is associated with temporary breakdowns of international borrowing and for how long this period lasts. Quantitatively I find that the model calibrated to other moments predicts consequences of default that are of empirically plausible magnitudes.

Of course the mechanisms modelled reflect only a part of a much more complicated reality. Yet the model offers insights that are relevant for a current policy debate on reducing banks' exposure to domestic sovereign debt. I show that any such policy not only makes banks more resilient against sovereign debt crisis; it also reduces the sustainability of sovereign debt.

Taking this thought further, this result also suggests that countries which have a higher share of domestically held debt and a more leveraged financial sector can sustain more debt. The development of leveraged financial sectors may hence explain the increase of debt to GDP ratios in the post-war period. Additional Supporting Information may be found in the online version of this article:.

The data and codes for this paper are available on the Journal website. They were checked for their ability to reproduce the results presented in the paper. Similar recent contributions are compared at the end of this section. Kletzer and Wright show how exclusion may arise as a renegotiation proof punishment strategy to make state-contingent trade sustainable in an environment of two-sided limited commitment. However, theirs is not a theory of exclusion since it never arises in equilibrium and its duration off-equilibrium is undetermined.

As one of the first papers to endogenise output losses Mendoza and Yue highlight a different, complementary mechanism: output drops because firms lose access to import goods markets. I add to these by also modelling financial frictions and exploring how they can lead to a persistent motive to borrow, endogenous default costs and exclusion. Exclusion is necessary to generate dynamic default costs in Mendoza and Yue , Mallucci , Engler and Grosse Steffen , Boz et al.

In Perez it is necessary for one of two types of default costs. The link from default to output losses through the financial sector is also explored in 3-period models like Basu , Bolton and Jeanne , Brutti , Mayer , Erce , Acharya et al. None of these models can speak about market exclusion or be tested quantitatively. Chari et al. For similar evidence, see De Paoli et al. In the few cases where the settlement periods do not coincide I follow GMR. In these cases I use the list in Schmitt-Grohe and Uribe For robustness I also consider the original GMR set, where I excluded those episodes that are not associated with banking crises.

Country FE are problematic due to the limited number of observations per country, since most countries do not reappear in the sample. I thus interpret these specifications as a robustness check. Due to insufficient sample size, results from the CT sample with country and year FE are omitted. For example, anticipation of reaccess might affect private credit positively, generating inverse causality bias.

Or there may be further omitted variables. This assumption is also in line with the finding that direct investment partially compensated for the reduction in intermediated credit during the financial crisis Becker and Ivashina, ; Fiore and Uhlig, For example, Gertler and Kiyotaki , who motivate an equivalent assumption by retiring bankers.

Sosa-Padilla and Gennaioli et al. Online Appendix D microfounds this constraint in a model where banks hold bonds for either liquidity or collateral reasons also see Perez, ; Engler and Grosse Steffen, Besides, notice that regulation and repression could be considered policy variables as in Chari et al. I abstract from this consideration. This is so because in equilibrium bonds pay a lower risk-adjusted return than loans see Subsection 3.

It implies that the distribution of bond holdings across foreign investors and banks is determined in the private equilibrium defined below. The possibility that bank equity might turn negative is ignored for simplicity. The model will later be calibrated such that this omission remains irrelevant. Assuming that banks carry some of the risk associated with firm investment would change the quantitative properties of the model, since it would introduce a positive correlation between the innovation in TFP and bank equity and would hence reduce the degree to which default incentives increase in TFP.

Lump sum taxation is a common simplifying assumption to rule out incentives for tax distortion smoothing. At the same time the model also fits other default episodes, such as the Argentinian and the Russian crises, where concerns regarding the damage to the financial sector also played a dominant role in the policy debate before and after default. To estimate the trend I use data only until , such that the estimate is not affected by the crisis. Since banks need bonds to operate, the government always will want to satisfy the domestic demand for bonds.

Here I abstract from this mechanical motive to issue bonds and focus on why the government has a motive to issue more bonds than domestic banks demand. The model also gives rise to other incentives that affect the optimal debt choice of the government.

The former motive results from risk aversion, the latter from the fact that foreign debt causes default, which is distortionary. I thank a referee for pointing this out.

For the calibration considered below, these two motives to save are outweighed by the return differential motive to borrow. Since models with exogenous output costs usually feature costs that increase in output and since default is triggered by a negative shock, the output costs increase over time after a typical default event e.

In contrast, in my model the costs increase over time even independently of any fundamental shock. Note that, together with the borrowing motive, this feature guarantees that default happens along the equilibrium path. The initial observations are discarded. The trade balance and the spread are not normalised. Episodes where another default happened in the 31 years prior to the default are excluded. As Hatchondo and Martinez and Chatterjee and Eyigungor show, higher ratios can be obtained with long-term debt, an extension beyond the scope of this paper for computational reasons.

However, they are also targeting higher default frequencies. As I explain in more detail there, what matters for the duration of exclusion is not the level of bank profits but their increase in times of crisis.

Thus a few unusual 1-period interruptions of exclusion due to unusually large shocks are ignored. A caveat: Schmitt-Grohe and Uribe report the duration until settlement, not until market reaccess. Gelos et al. While average durations differ across studies, the skewness of their distribution is a robust feature.

However, unlike in the data, in the model the recovery of credit to GDP is not yet complete at time of reaccess. This is because reaccess happens around the time that banks become the marginal investors. At this time bank equity has recovered only partially yet.

This implies that the deposit-loan spread is still elevated, causing an elevated marginal product of capital MPK in turn. The penultimate row of Table 3 reports the trough of GDP after default, when the economy starts from the risk adjusted steady state, as in Figure 2.

This measure is unaffected by the change in the default frequency. A caveat: if banks' bias towards domestic public debt reflects unmodelled optimal behaviour, then forcing them to hold less domestic bonds may have further negative consequences beyond the scope of this paper. Acharya V.

Google Scholar. Aguiar M. Aiyagari S. Arellano C. Balke N. Balteanu I. Basu S. Becker B. Benhima K. Benjamin D. Bocola L. Bolton P. Boz E. Broner F. Brunnermeier M. Brutti F. Chamberlain G. Chari V. Chatterjee S. Cruces J. De Paoli B. Demirguc-Kunt A. Eaton J. Engler P. Erce A. Fiore F. Gelos R. Gennaioli N. Gertler M. Google Preview. Gordon G. Gornemann N. Gourinchas P. Greenwood J.

Guimaraes B. Hatchondo J. Holmstrom B. Joo H. Kiyotaki N. Klein P. Kletzer K. Kumhof M. Mallucci E. Mayer M. Mendoza E.

Mengus E. Park J. Perez D. Perry G. Reinhart C. Richmond C. Schmitt-Grohe S. Sosa-Padilla C. Sturzenegger F. Svirydzenka K. Oxford University Press is a department of the University of Oxford.

It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide. Sign In or Create an Account. Sign In. Advanced Search. Search Menu. Article Navigation. Close mobile search navigation Article Navigation. Volume Dominik Thaler Dominik Thaler. Email: dominik. Oxford Academic. Select Format Select format. Permissions Icon Permissions. Abstract Why do governments borrow internationally? Open in new tab Download slide.

To corroborate this finding, I next estimate the effect of financial sector conditions on the probability of an exclusion period ending. In order to address potential reverse causality issues, I lag my explanatory variables. Table 1. Determinants of Market Reaccess. Open in new tab. There is a continuum of mass 1 of identical households, which each make a consumption—labour—savings choice.

Both assets are safe and loans cannot be shorted. The household has rational expectations and chooses labour, consumption and investment to maximise its lifetime utility, taking prices and aggregate states as given. There is continuum of mass 1 of banks, each run by a banker. Each banker starts the period with a portfolio of 1-period assets and liabilities chosen last period. The first constraint is crucial. Following Kiyotaki and Moore I assume that bankers face a commitment problem: at the end of a period after investing bankers can repudiate their deposit liabilities.

This gives the banker the opportunity to renegotiate his liabilities. Depositors anticipate this and will not let the value of deposits exceed the value of recoverable assets. The banker is a small member of the household, to whom he pays the dividend.

He maximises the discounted value of dividends, taking the household discount factor, prices and aggregate state as given. There is a continuum of mass 1 of firms that solve a 2-period problem. The rest of the world is represented by perfectly competitive risk-neutral deep-pocket foreign investors. Lending to domestic private agents requires local know-how and is hence not possible or profitable for foreign investors, but they can invest in government bonds. Assuming this type of market segmentation reasonably approximates many less financially integrated economies, where the total net foreign asset position is almost entirely due to sovereign debt.

This assumption will be crucial in driving sovereign debt dynamics. Having discussed all private agents, we are now ready to close the private economy by the corresponding market clearing conditions.

I denote aggregate choice variables by capital letters, i. Definition 1 Private equilibrium. It is important to note that the government has rational expectations and —unlike the private agents—is non-atomistic. This means that it understands how its own decisions influence the choices and expectations of households, banks, firms and foreign lenders and the resulting market prices both today and in the future. Furthermore, the government understands its own commitment problem and correctly anticipates the behaviour of the government in the future.

Definition 2 Full equilibrium. Post-Default Dynamics. I need help. More details. Type of publication: Article Notes: Published. Saved in favorites. Similar items by subject. Similar items by person. A service of the. Sitemap Contact us Imprint Privacy. Guimaraes, Bernardo. Sovereign debt Default World interest rates Output shocks.

Save to Library Save. Create Alert Alert. Share This Paper. Supplemental Presentations. Presentation Slides. Sovereign default: which shocks matter? Figures from this paper. Citation Type. Has PDF. Publication Type. More Filters. Sovereign default risk and commitment for fiscal adjustment. This paper studies fiscal policy in a model of sovereign debt and default.

A time inconsistency problem arises: since the price of past debt cannot be affected by current fiscal policy and … Expand. A decline in international risk-free interest rates decreases emerging markets EM sovereign spreads. I show that a quantitative model of sovereign debt and default exhibits this pattern if foreign … Expand. View 2 excerpts, cites background. This paper develops a two-country asset pricing model with defaultable firms and governments.

This model shows that higher sovereign credit risk in a country depresses equity prices internationally … Expand. This paper builds a unified model of sovereign debt, default risk, and news shocks. News shocks improve the quantitative performance of the sovereign default model in a number of empirically-relevant … Expand.

This paper builds a model of sovereign debt in which default risk, interest rates, and debt depend not only on current fundamentals but also on news about future fundamentals. News shocks affect … Expand. View 1 excerpt. Borrowing from thy neighbour: a European perspective on sovereign debt.

European capital markets show increasing concern about the extent of sovereign debts and their sustainability.