Academic journal article Economic Inquiry

Fiscal Rules and Government Borrowing Costs: International Evidence

Academic journal article Economic Inquiry

Fiscal Rules and Government Borrowing Costs: International Evidence

Article excerpt

I. INTRODUCTION

Numerical fiscal rules have been a popular addition to fiscal frameworks since the early 1990s. (1) In this paper, we examine whether the adoption of such rules has had a beneficial impact on government borrowing costs in international and domestic credit markets. Given the considerable executive and legislative effort involved in the adoption of rules, whether or not their adoption has impacted on borrowing costs is pertinent. The main reason for believing that they might do so is if numerical rules add to fiscal policy credibility and reduce the risk premium on government debt. If they do, we would expect this to be reflected in lower borrowing costs in credit markets for governments that have adopted a rule compared to the costs for governments that have not adopted a rule. In addition, we would expect government borrowing costs to decline even with respect to other (private sector) borrowers more generally. We examine these propositions in the context of developments in "spreads" for government borrowing in international and domestic credit markets.

In the international market, government borrowing costs are represented by the spread between the rate at which a country borrows and the "risk free" interest rate, defined as the yield on long-term (10-year) U.S. Treasury bonds. In this market, the interest rate paid by governments is typically higher than the yield on U.S. bonds. If the adoption of fiscal rules adds to policy credibility then, ceteris paribus, we would expect the yield the government would need to offer on its bonds to decline relative to the U.S. yield--that is, the spread on borrowing in the international market would decline. In the domestic market, we focus on the spread between the interest rate charged by domestic banks on loans to private sector entities and the interest rate at which governments can borrow through the issuance of short-term securities. In this market, the cost of borrowing by the government is typically lower than it is for private sector entities. Accordingly, if adopting fiscal rules reduces the risk premium on government borrowing, we would expect this spread to widen as the cost of borrowing by the government declines relative to that of private sector entities.

Our paper makes two contributions to the literature on fiscal rules and borrowing costs. First, we deal with an important econometric issue in evaluating the effect of fiscal rules, which is the nonrandom selection of policy options that arises when a country's fiscal policy regime choice is systematically correlated with a set of observable variables that also affect the outcomes and that can lead to biased estimates. Many of the studies of the impact of fiscal rules have failed to address this issue. We address it by evaluating the treatment effect of numerical fiscal rules on borrowing costs in a panel dataset of countries, many of which have adopted numerical fiscal rules in recent years. Second, the little empirical evidence that has been published on the impact of fiscal rules on borrowing costs relates mainly to the experience of U.S. states and European Union countries. In contrast, our dataset contains annual observations for 101 advanced and developing countries for the period 1985-2012, of which 44 countries adopted a numerical rule on the fiscal balance and/or the stock of public debt.

To anticipate our baseline results, we find the average treatment effect on government borrowing costs of adopting a fiscal rule to be strong and robust, leading on average to a reduction in the relative cost of borrowing by the government of between 1.1% and 1.8% of the borrowing spread in international credit markets and between 1.2% and 1.9% of the borrowing spread in domestic credit markets. Those countries that have already adopted fiscal rules, or are considering adopting them, should find our results encouraging.

The rest of the paper is organized as follows. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.