بدهی، نرخ بهره، و ادغام بازارهای مالی

It is commonly believed that higher budget deficits raise interest rates. However, these crowding out effects of increasing public debt have usually been found to be small or non-existent. One explanation is that on globalised bond markets interest rate differentials are offset due to financial integration. This paper tests crowding out, and measures the degree of integration of government bond markets, using spatial modelling techniques. Our main finding is that the crowding out effect of public debt on domestic long term interest rates is small: a 1% increase in the debt ratio pushes up domestic rates by 2 pp at most. Financial integration implies an important spillover effect via international bond markets, but only between OECD, and in particular EU, countries. The feedback effect from these markets on long term interest rates is as important as the domestic crowding out effect of higher public debt. Emerging markets are not as well integrated into international capital markets, causing a stronger crowding out effect.

A government running a deficit needs to turn to financial markets to place additional public debt. This supply of newly issued public bonds competes for financing with bonds issued by firms, and so pushes up long term interest rates. Since higher rates crowd out private investment, institutions like the IMF or OECD recommend consolidating public finances to harness economic growth. Despite being widely accepted in the economics profession as an important effect of public debt, there is surprisingly little robust empirical support for crowding out (Engen and Hubbard, 2004). A first explanation is that economic agents anticipate paying down currently high deficits with higher taxes in the future. Under Ricardian Equivalence, private saving fully offsets the effect of a higher deficit. However, there is by now a large body of empirical evidence that clearly refutes the zero impact of deficits on aggregate macroeconomic variables (Chung and Leeper, 2007). A second rationale for a lack of crowding out is capital mobility. Domestic and foreign agents diversify their holdings across borders, including also government bonds to their portfolio. Capital flows offset any interest rate differential following an increase in the domestic supply of government bonds. Under full capital mobility, domestic interest rates rise in step with global rates, and the crowding out effect is infinitesimally small. In practice, capital mobility is far from complete as foreign and domestic assets are imperfect substitutes due to incomplete information or risk aversion. Interest rate differentials persist when the spillover between markets is weak. The typical empirical test for crowding out regresses a domestic interest rate on domestic public debt, and controls for spillover including proxy measures of capital mobility, such as aggregate capital flows or a composite measure of foreign debt or foreign interest rates. This supposes an identical transmission of fiscal policy across financial markets, and the focus usually is on a particular subset of OECD countries. However, the interactions between bond markets are much more complex in reality. The spillover works out on global financial markets via various channels, and affects more strongly countries that are more closely integrated. The spillover is contemporaneous for a large group of countries and feeds back to the domestic bond market. In this paper, we follow standard practice in testing crowding out and explain nominal long term interest rates by public debt. But we control for the spillover by adding a spatial term that captures the degree of financial integration. This term models the contemporaneous co-movement of domestic and foreign interest rates, and so quantifies in a straightforward way the degree of integration of government bond markets. The spatial measure then allows calculating the general equilibrium effect of higher public debt on interest rates, taking into account the spillover to other markets and the feedback to the domestic bond market. We use data for a panel of both OECD and emerging market economies over the period 1990–2005. Our main finding is that the domestic crowding effect of public debt is small but significant. It adds at most 2 basis points for every percentage rise in the debt to GDP ratio. As the correlation between global bond market returns is just 0.10, there is limited spillover on international bond markets. The reason is that emerging markets are not fully integrated in global bond markets. By contrast, we find bond markets to be more strongly integrated among OECD, and especially EU, countries. The feedback effect from these markets on long term interest rates is as important as the domestic crowding out effect of higher public debt. Spillover reflects deep economic integration but also economic co-movement in major crisis moments. Various measures of cross-country linkages give broadly similar results. The findings are also robust to alternative specifications and data definitions. The paper is structured as follows. In Section 2, we discuss a simple theoretical model for testing crowding out and measuring the effects of financial integration. We then continue in Section 3 by discussing the results of the spatial panel model, and provide several robustness checks in Section 4. The final section summarises the main results, and discusses some policy implications.

There is much discussion about the effect of fiscal expansions on interest rates. A lack of response of interest rates can be justified under two different theoretical conditions. First, under Ricardian Equivalence, deficits do not affect macroeconomic variables as economic agents anticipate the paydown of higher deficits with future taxes. Second, capital flows between economically integrated economies offset any interest rate differentials that follow upon an increase in the supply of government bonds. Fiscal deficits are not necessarily financed by domestic financial resources only. In this paper, we extend a simple empirical model for testing crowding out and apply spatial panel techniques. Spatial models impose few restrictions on the spillover, as all contemporaneous interactions on capital markets in many countries are taken into account. This co-movement of interest rates on financial markets in all nearby foreign economies is assumed not to spread symmetrically across borders. This correlation of market returns gives an easily interpretable measure of the degree of integration of government bond markets, which lies between 0 and 1. We test the effect of financial integration on crowding out for a panel of OECD and emerging economies over the period 1990–2005. Our main finding is that the crowding out effect of public debt on domestic long term interest rates is small. A 1% increase in the debt ratio pushes up domestic rates by 2 pp at most. Financial integration implies limited spillover via financial markets. Cross border spillover is much stronger among OECD, and in particular EU, countries. Emerging markets are not fully integrated in global markets. Spillover reflects deep economic integration but also economic co-movement. Our main result is robust to various checks. These results have some implications for fiscal policy. Persistent increases in deficits lead to large accumulated effects over time (Friedman, 2005), and in crisis periods, debt often rises by double digit numbers. The argument for coordination of fiscal policy is not convincing in case the spillover occurs on capital markets. After all, the mitigating effect of financial markets is a purely pecuniary externality and does not require international coordination. The allocation of savings to the public or private sector, whether at home or abroad, is efficient. But as financial globalisation gives access to cheap international financing, market discipline may not be sufficient to keep debt under control. In case spillover is related to contagion on financial markets (in the case of emerging economies) or to monetary union (in the case of EMU), some mechanisms might be necessary to correct the distortion on capital markets.