Skip to main content
Log in

Geographic proximity, long-term institutional ownership, and corporate social responsibility

  • Original Research
  • Published:
Review of Quantitative Finance and Accounting Aims and scope Submit manuscript

Abstract

Building on the premises that institutional ownerships vary in their impact on corporate social responsibility (CSR) decisions and that geographic proximity facilitates the valuation of benefits from CSR, we hypothesize that local long-term institutional ownership is a driver for corporate social performance, in particular positive CSR (CSR strengths). Using a panel data of S&P 500 firms over a 15-year window, we show that long-term institutional ownership that varies in geographic proximity to the focal firm does have a heterogeneous impact on CSR. Whereas both local and non-local long-term institutional ownership has a similar negative effect on CSR concerns, only local long-term institutional ownership has a positive effect on CSR strengths. The positive relation between local long-term institutional ownership and CSR is stronger in firms that are more involved in dealing with soft information which is difficult to quantify from a distance, such as those with high levels of research and development and intangible assets.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Code availability

Available upon request.

Notes

  1. Financial (SIC 6000-6999) and regulated utility firms (SIC 4900-4999) are not included in our sample.

  2. KLD used ticker as an identifier for the firms it covered prior to 1995 and switched to CUSIP as firm identifiers since 1995. To minimize the possibility of misidentified firms when combining data with Compustat, which uses CUSIP as firm identifiers, we work with data starting from 1995. In 2010, KLD made significant changes caused by the change in KLD ownership. As a result, data gathered before and after 2010 data are not fully compatible. For this reason, we limit our data from 1995 to 2009, just as Harrison and Berman did (2016).

  3. The reported positions are those in which the institution owns more than 10,000 shares or with over $200,000 in market value.

  4. Bushee (2001) kindly provides the institutional investor classification data (1981–2009) on his website:

    http://acct3.wharton.upenn.edu/faculty/bushee/.

  5. The seven areas are environment, community, corporate governance, diversity, employee relations, human rights, product quality and safety.

  6. We define High10 with an above-10% threshold to identify firms in areas with high concentration of financial institutions, our results continue to hold and are similar to what we report in the paper with the threshold being 11% and 12%.

  7. Index funds belong to long-term institutional ownership because of their long investment time horizon.

  8. We thank Bin Wang for providing the data set.

  9. We also construct another sample including all firms after adjusting for any changed headquarters and re-estimate the relation between LLTIO and CSR. All the results are available upon request.

  10. These robustness checks include alternative definition of CSP (Choi and Wang 2009) and LLTIOs (within-state and 250-mile distance), subsample analysis (pre- and post-2002 Sarbanes–Sox periods), and controlling for time-varying industry fixed effects. We do not include these results to save space.

References

Download references

Funding

Not applicable.

Author information

Authors and Affiliations

Authors

Contributions

Not applicable

Corresponding author

Correspondence to Ying Li.

Ethics declarations

Conflict of interest

Not applicable.

Availability of data and transparency

Public and commercial data.

Ethics approval

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix 1: Definitions of variables

Variable

Name

Variable definition

CSR related

CSR strength scores

CSR STR

Sum of adjusted Community, Diversity, Employee Relations, Environment, and Product strength scores (Source: KLD database); how adjusted strength scores are calculated are given in “Appendix 2

CSR concern scores

CSR CON

Sum of adjusted Community, Diversity, Employee Relations, Environment, and Product concern scores (Source: KLD database); how adjusted concerning scores are calculated are given in “Appendix 2

Net CSR scores

NET CSR

Sum of net adjusted Community, Diversity, Employee Relations, Environment, and Product strength scores (Source: KLD database); how adjusted net strength scores are calculated are given in “Appendix 2

Finance variables

Log (total assets)

LogTA

Log of total assets [at] (source: Compustat)

Leverage

Leverage

Debt to total asset ratio [(dltt + dlcc)/at] (source: Compustat)

Return on asset

ROA

Net income divided by total assets [(ni/at)] (source: Compustat)

Tobin’s Q

Q

Market value of assets divided by book value of assets [(prcc_f*csho + at-ceq)/at](source:Compustat)

Fixed asset ratio

FA/TA

Fixed assets divided by total assets [ppent/at](source: Compustat)

R&D intensity

R&D/TA

R&D expense divided by total assets [xrd/at]; missing R&D expense is treated as zero (source:Compustat)

Advertising intensity

ADV/TA

Advertising expense divided by total assets [xad/at]; missiong advertising expense is treated as zero (source: Compustat)

Log (firm age)

Logage

Log (1 + firm age); firm age is measured as fiscal year minus the first year that the firm is appeared in Compustat

Institutions related

Total institutional ownership

IOR

Total institutional ownership for a firm in a given fiscal year; Total institutional shares/total number of share outstanding (source: Thomson Reuters’ 13F)

Long-term institutional ownership

LTIO

Long-term institutional share/total number of shares outstanding; Dedicated and Quasi institutions are treated as long-term institutions (source: Thomson Reuters’ 13 F & Professor Brian Bushee’s web site: http://acct3.wharton.upenn.edu/faculty/bushee/)

Dedicated institutional ownership

DIO

Dedicated institutional share/total number of shares outstanding; (source: Thomson Reuters’ 13 F & Professor Brian Bushee’s web site: http://acct3.wharton.upenn.edu/faculty/bushee/)

Quasi-indexer institutional ownership

QIO

Quasi-indexer institutional share/total number of shares outstanding; (source: Thomson Reuters’ 13 F & Professor Brian Bushee’s web site: http://acct3.wharton.upenn.edu/faculty/bushee/)

Local institutional ownership

Localown

Local institutional shares/total number of shares outstanding; Institutions are defined as “local” if the distance between the firm’s and the institution’s headquarters is 100 miles or less (source: Compustat; Thomson Reuters’ 13 F)

Local long-term institutional ownership

LLTIO

Local long-term institutional shares/total number of shares outstanding; Institutions are defined as “local” if the distance between the firm’s and the institution’s headquarters is 100 miles or less (source: Compustat; Thomson Reuters’ 13 F)

Local dedicated institutional ownership

LDIO

Local dedicated institutional shares/total number of shares outstanding; Institutions are defined as “local” if the distance between the firm’s and the institution’s headquarters is 100 miles or less (source: Compustat; Thomson Reuters’ 13 F)

Local quasi-indexer institutional ownership

LQIO

Local quasi-indexer institutional shares/total number of shares outstanding; Institutions are defined as “local” if the distance between the firm’s and the institution’s headquarters is 100 miles or less (source: Compustat; Thomson Reuters’ 13 F)

Non-local long-term institutional ownership

NLLTIO

Non-local long-term institutional shares/total number of shares outstanding; Institutions are defined as “local” if the distance between the firm’s and the institution’s headquarters is 100 miles or less (source: Compustat; Thomson Reuters’ 13 F)

High10*LLTIO

High10 * LLTIO

High10 is 1 if LLTIO is 10% or higher, else zero. High10 multiplied by local long-term institutional ownership (LLTIO) (source: Compustat; Thomson Reuters’ 13 F)

Appendix 2: Detailed definitions of KLD CSR variables

We follow definitions in Jo and Harjoto (2011, 2012) with a slight change.

CSR strength and concern scores

Com Str(Con) (i, t) = [sum of all community strength (concern) score for firm i at year t divided by maximum number of community strength (concern) score during year t]

Div Str(Con) (i, t) = [sum of all diversity strength (concern) score for firm i at year t divided by maximum number of diversity strength (concern) score during year t]

Emp Str(Con) (i, t) = [sum of all employee relations strength (concern) score for firm i at year t divided by maximum number of employee relations strength (concern) score during year t]

Env Str(Con) (i, t) = [sum of all environment strength (concern) score for firm i at year t divided by maximum number of environment strength (concern) score during year t]

Pro Str(Con) (i, t) = [sum of all product strength (concern) score for firm i at year t divided by maximum number of product strength (concern) score during year t]

CSR STR(CON) (i, t) = Com str(con)(i, t)+ Div str(con) (i, t)+ Emp str(con)(i, t) + Env str(con) (i, t)+ Pro str(con) (i, t); combined CSR strength (concern) score

Net CSR scores

Net Com(i, t) = [(sum of all community strength score for firm i at year t) − (sum of all community concern scores for firm i at year t)]/[maximum score of community strength at year(t) + maximum score of community concern at year (t)]

Net Div(i, t) = [(sum of all diversity strength score for firm i at year t) − (sum of all diversity concern scores for firm i at year t)]/[maximum score of diversity strength at year t + maximum score of diversity concern at year t]

Net Emp(i, t) = [(sum of all employee relations strength score for firm i at year t) − (sum of all employee relations concern scores for firm i at year t)]/[maximum score of employee relations strength at year t + maximum score of employee relations concern at year t]

Net Env(i, t) = [(sum of all environment strength score for firm i at year t) − (sum of all environment concern scores for firm i at year t)]/[maximum score of environment strength at year t + maximum score of environment concern at year t]

Net Pro(i, t) = [(sum of all product strength score for firm(i) at year(t)) − (sum of all product concern scores for firm(i) at year(t))]/[maximum score of product strength at year(t) + maximum score of product concern at year(t)]

NET CSR(i, t) = Net Com(i, t) + Net Div(i, t) + Net Emp(i, t) + Net Env(i, t) + Net Pro(i, t); combined Net CSR score

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chang, K., Kabongo, J. & Li, Y. Geographic proximity, long-term institutional ownership, and corporate social responsibility. Rev Quant Finan Acc 56, 297–328 (2021). https://doi.org/10.1007/s11156-020-00895-9

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11156-020-00895-9

Keywords

JEL Classification

Navigation