Bank dependence in emerging countries: Cross-border information percolation in mutual fund equity investing

Abstract

In contrast to the literature involving U.S. bank domestic lending, we find that mutual funds affiliated with lending banks reduce their equity investment and turnover in the non-U.S.-listed stock of their non-U.S. borrowers compared to non-lending banks or unaffiliated mutual funds. Reduced equity holdings increase loan spreads, preserving the lending bank’s cross-border information monopoly. Equity market holdings and turnover are reduced when banks lend to firms in emerging nations and when the geographic distance between the lender and the mutual fund manager is greatest. Thereby, long-range information percolation may benefit global institutions at the expense of individual subsidiaries.

Résumé

Contrairement à la litérature sur les prêts consentis par les banques américaines sur le marché intérieur, nous constatons que les fonds communs de placement affiliés à des banques prêteuses réduisent leurs placements et leurs renouvellements en actions pour les titres non américains cotés en bourse de leurs emprunteurs non américains comparativement aux banques non prêteuses ou aux fonds communs de placement non affiliés. La réduction des participations en actions augmente les écarts de taux sur les prêts, préservant ainsi le monopole transfrontalier de la banque prêteuse en matière d’information. Les placements et les renouvellements sur les marchés boursiers sont réduits lorsque les banques prêtent aux entreprises des pays émergents et que la distance géographique entre le prêteur et le gestionnaire de fonds communs de placement est la plus grande. Ainsi, la diffusion à long terme de l’information peut profiter aux institutions mondiales au détriment des filiales individuelles.

Resumen

En contraste con la literatura sobre préstamos nacionales de bancos de los Estados Unidos, encontramos que los fondos mutuos afiliados a bancos prestamistas reducen su inversión de capital y su rotación los prestamistas que no están listados en la bolsa de valores de los Estados Unidos comparado con los bancos que no son prestamistas o que no están afiliados a fondos mutuales. Las tenencias reducidas de capital aumentan la propagación de préstamos, preservando el monopolio de información transfronterizo del banco prestamista. Las tenencias y rotación del mercado bursátil son reducidas cuando los bancos prestan a empresas en naciones emergentes y cuando la distancia geográfica entre el prestamista y el gerente fondo mutual es mayor. Así, la filtración de información a distancia puede beneficiar a instituciones globales a expensas de las filiares individuales.

Resumo

Em contraste com a literatura acerca de empréstimos domésticos bancários nos Estados Unidos, descobrimos que fundos mútuos afiliados a bancos emprestadores reduzem investimento em capital e turnover em ações listadas fora dos EUA de seus mutuários não norteamericanos em comparação com bancos não emprestadores ou fundos mútuos não afiliados. Participações acionárias reduzidas aumentam spreads de empréstimos, preservando o monopólio de informações transfronteiriças do banco emprestador. Participações no mercado de ações e turnover são reduzidos quando bancos emprestam para empresas em países emergentes e quando a distância geográfica entre o emprestador e o administrador do fundo mútuo é maior. Assim, a filtragem de informações de longo alcance pode beneficiar instituições globais em detrimento de subsidiárias individuais.

摘要

与涉及美国银行国内贷款的文献相比,我们发现与非贷款银行或非附属的共同基金相比, 附属于贷款银行的共同基金减少了非美国借贷人的非美国上市股票的股权投资和营业额。减持的股权增加了贷款利差, 维持了贷款银行的跨境信息垄断。当银行向新兴国家公司提供贷款和当贷方与共同基金经理人之间的地理距离最大时, 股票市场持有量和营业额会减少。因此, 远程信息渗滤可能会以个体子公司作为代价而使全球机构受益。

Appendix

Data Appendix: detailed description of variables

Variable name	Definitions
Dependent variables
ΔLOANSij,t	LOANSij,t–LOANSij,t−1, where LOANSij,t is the dollar value of all newly granted loans to company i in quarter t by bank j as a percent of total syndicated bank loans initiated by bank j in quarter t
ΔEQUITYij,t	EQUITYij,t–EQUITYij,t−1, where EQUITYij,t is the dollar value of mutual fund holdings in company i by parent j divided by the total dollar amount of mutual fund holdings of parent j in quarter t that contains the deal active date
Loan_spreadij,t	The weighted average all-in-spread drawn for the maximum facility amount in each loan deal for company i by bank j in quarter t
RELTURNij,t	ln(1 + TURNij,t/$\sum {\text{TURN}}_{ij,t}$), where TURNij,t is the normalized (as a fraction of total mutual fund turnover) average turnover (absolute value of quarterly trading) by parent j in company i’s equity holdings in the loan origination quarter t
Main explanatory variables
Lending conglomerate dummy	Equal to one if the mutual funds belong to a parent j that is affiliated to a bank lending to firm i in a particular deal, and zero otherwise. This variable corresponds to the main explanatory variable in Massa and Rehman (2008)
Affiliation dummy	Equal to one if the mutual funds belong to a parent j affiliated with any bank, and zero otherwise
No prior equity dummy	Equal to one if parent j’s mutual funds have no equity holdings in firm i during the three quarters prior to loan origination, and zero otherwise
Emerging market dummy	Equal to one for any country that is classified as “emerging” under MSCI-Barra classification, and zero otherwise
Financial development	A measure of Capital Account openness of a country calculated by Chinn and Ito (2006) and updated to 2016
Distance	Distance is measured as Euclidean distance between the country capital city of mutual fund parent j’s headquarters and the country capital of the lending bank in order
Control variables
Lagged MSCI return	Quarterly return on the MSCI stock index for the country where company i is headquartered in the quarter (t−1) before the deal
Lnassets	Log of the total assets of borrowing/equity issuing company i in the year ending before the deal active date
Cash	Cash and cash equivalents deflated by total assets of company i in the year ending before the deal active date
Return on assets = Roa	Operating income before depreciation deflated by total assets of company i in the year ending before the deal active date
Market to book = m2b	Sum of the total book value of debt and the market value of equity deflated by the firm’s total assets of company i in the year ending before the deal active date
Lnparenttotassets	Log of the total assets of parent j
Probability	Probability that a bank chooses to be in the mutual fund investing business as opposed to just being a lender. This is the output from the first stage of our multinomial logistic setup that is included in the second stage to control for sample selection bias
Interest income	Interest income divided by total income calculated at the bank holding company level