A large literature in macroeconomics argues that general equilibrium effects matter for the transmission of fiscal policy shocks. For example, government spending or taxation changes (fiscal policy) have been shown to spill over through channels such as production factor (labor, capital, etc.) reallocation, output and input price changes, household consumption and saving responses, and monetary policy responses. However, despite the emergence of a burgeoning new macro-finance literature, there is limited work on whether and how fiscal policy shocks can be transmitted through the financial system to affect other parts of the economy.
In our recent paper, we propose and test a new financial intermediation channel of fiscal policy transmission. In our proposed channel, some firms in the economy directly benefit from a fiscal policy change (say, a tax cut). Affected firms then transmit the fiscal policy change to their lenders via reductions in credit risk. Finally, lenders transmit the fiscal policy change to otherwise unaffected borrowers through changes in the supply of credit. Simply put, some borrowers are made less risky by tax cuts, banks internalize this change, and lend in greater amounts to other firms. We refer to this channel as the credit channel of fiscal policy transmission.
The fiscal policy change that we use to uncover this channel is the 2004 American Jobs Creation Act (“AJCA”). The AJCA is special because it was largely unanticipated–making it a plausible shock to unsuspecting firms–and so was likely unrelated to investment opportunities for both targeted and untargeted firms. Furthermore, it temporarily reduced the taxes owed by U.S. multinational firms by cutting taxation on foreign income repatriated to the United States in 2004 and 2005. This temporary holiday created a large increase in the amount of repatriated foreign earnings, allowing for the measurement of effects on firms. Because the AJCA only affected multinationals, a subset of U.S. firms, spillovers to other, purely domestic firms can be isolated as they are not directly affected by the policy. In contrast, other pieces of fiscal legislation like the 2009 American Recovery and Reinvestment Act and the 2017 Tax Cuts and Jobs Act contained a number of different policy changes affecting large swaths of the economy. Such entanglements exhibited by other policies make it difficult to empirically distinguish the direct and indirect effects of the policy change.
We find strong support for the existence of a credit channel of fiscal policy transmission. Lenders significantly increase credit supply during the AJCA tax holiday period if they have larger exposures to AJCA-eligible firms. We find this is true after controlling for heterogeneity in lender-borrower matches as well as the demand for loans by individual borrowers. That is, lenders with greater exposure to AJCA-eligible firms make a disproportionately larger share of loans during this period, controlling for the amount firms borrow. In any design like this, one could worry about confounding differences in pre-trends but we find no differences in lending volumes between high- and low-exposure lenders prior to the tax holiday in 2004. In addition, the spike in credit supply associated with exposure coincides exactly with the beginning of the tax holiday period and ends immediately after the temporary holiday ends. Quantitatively, we find that a one standard deviation increase in exposure is associated with a 1.4% increase in bank-level credit supply during the tax holiday. Commercial banks (rather than nonbank lenders) account for the entirety of the increase. Loan terms also improve, consistent with a supply channel: loan amounts are greater; spreads are lower; maturities are longer; loans are more likely to be unsecured by collateral; and loans are more likely to include a revolving credit facility when they are originated during the tax holiday by lenders with a high level of existing exposure to AJCA-eligible borrowers.
We also find that firms receiving additional credit–and in particular, purely domestic firms without foreign income to repatriate–significantly increase investment following the passage of the AJCA. Since borrowers’ investment opportunities (and hence, their demand for credit) are endogenous, we instrument for credit supply using a borrower’s pre-AJCA exposure to high-exposure lenders. We see increases in capital expenditures, R&D, and acquisitions, in sum a $0.13 increase in investment for every dollar of additional lending they receive. Hence, the AJCA–designed in part to increase domestic investment by multinational firms–led to a material increase in investment by purely domestic firms as a result of the financial-sector spillovers we document.
We next attempt to identify the channel(s) through which the AJCA spilled over through the financial system to affect other firms in the economy. First, Oler, Shevlin, and Wilson (2007) find that the market value of repatriating firms increases as a result of the tax holiday. This increase in value reduces the risk of lending to such firms, thereby freeing up creditors’ capital. Second, multinational firms may use repatriated earnings to pay down debt (or otherwise reduce their demand for loans), thereby freeing up capital that lenders can lend to other borrowers. Finally, it is possible that exposed lenders are relatively better informed about the effects of the AJCA through their lending relationships with affected borrowers.
Our results are most consistent with the first mechanism. We find that market values increase and market leverage decreases for multinational borrowers around the passage of the AJCA, thereby making existing loans to such borrowers arguably less risky. And in fact we find that defaults on such loans drop significantly after the commencement of the tax holiday, suggesting that the riskiness of existing loans to multinational borrowers declined following the announcement of the tax holiday. In contrast, we find little evidence that multinational firms reduced their demand for credit or used repatriated earnings to pay down their existing loans. We also find larger effects for purely domestic borrowers than for the multinational borrowers who would be the most likely beneficiaries of having informed lenders.
Our findings have three important implications. First, they provide evidence of a new type of general equilibrium fiscal policy spillover in the form of credit reallocation. Second, they show that the effects of the AJCA went far beyond the multinational firms that were the direct beneficiaries of the policy and which have been the subject of the existing literature on the AJCA and numerous reports from the Joint Committee on Taxation and the Congressional Budget Office. Finally, we present the first results suggesting that borrower financial shocks can be transmitted through the asset side of banks’ balance sheets.
Andrew Bird is an Assistant Professor at the George L. Argyros School of Business and Economics of Chapman University.
This post is adapted from his paper, “The Credit Channel of Fiscal Policy Transmission,” available on SSRN and co-authored by Stephen A. Karolyi, Stefan Lewellen, and Thomas G. Ruchti.
