hazard present in the firm and enable a firm to better extract benefits from its human capital. Thus, the need
for the additional information captured in non-financial measures will not be as great for firms that rely
on human capital in an egalitarian environment as opposed to a hierarchical environment. In conclusion,
it is argued that the pay structure will moderate the relation between the reliance on human capital and
the use of non-financial measures. This leads to the following hypotheses:
H2a. The pay structure moderates the relation between the use of non-financial measures and the reliance
on human capital. Specifically, the likelihood of the use of non-financial measures in bonus compensation
for firms that rely on human capital is greater when firms use hierarchical pay structures than when they
use egalitarian pay structures.
H2b. The pay structure moderates the relation between the use of human resource measures and the
reliance on human capital. Specifically, the likelihood of the use of human resource measures in bonus
compensation for firms that rely on human capital is greater when firms use hierarchical pay structures
than when they use egalitarian pay structures.
3. Research methods
3.1. Sample
I limit the sample of firms to those that report in no more than five four-digit SIC codes in the same
division code.7 Firms are required to have more than 250 employees and report sales, total assets, total
liabilities, common shares outstanding, closing price, the number of employees, and labor costs.8 This
brings the sample to 1662 firms prior to the restriction on labor costs. Requiring firms to have labor costs
further reduces the sample to 312 firms. In addition to the reduction in sample size, this restriction shifts
the sample from an emphasis on manufacturing firms to one on service-type of firms (e.g., transportation,
communication, financial services, etc.). I also require that the firm have an executive bonus program
and that the 1997 proxy statement, inclusive of the compensation committee report, be available and
interpretable9 on Compact Disclosure. This results in a final sample of 177 firms. Descriptive data are
found in Table 1.
problem. An example of a horizon problem is when managers do not reduce production when demand falls. If managers are
evaluated on margin and the production process uses full cost accounting, then a focus on financial incentives may induce
managers to maintain (or increase) production, thus forsaking the strategic objectives of the organization in order to meet
financial outcomes. Focusing on non-financial measures, especially human resource measures, directly related to the firm’s
strategy could help mitigate the potential adverse behaviors.
7 Highly diversified firms may pursue multiple strategies and rely on multiple resources that differ across business segments,
which would add considerable noise to this study.
8 Labor and related expenses is data field #A42. Compustat defines this item as “salaries, wages, pension costs, profit sharing
and incentive compensation, payroll taxes, and other employee benefits.” According to a representative from Compustat, this
data field does not articulate with a firm’s income statement; rather, it represents the total amount spent on labor and related
expenses regardless of how the cost is accounted for. This is neither a required data item nor is this a required disclosure in the
financial statements. Therefore, this study uses firms that self-select into a group whose members choose to provide this item in
their public documents. This restriction could have the effect of reducing the variability of the labor intensity ratios since these
firms probably rely more on labor than do firms that do not disclose labor costs.
9 The compensation committee report had to contain enough information to support an accurate coding of the variables.
(1) Firms are selected using the Compustat Industrial Files and the Compustat files. Firms report in five or fewer SIC codes
within one division code. Must report sales, have employees >250 and depreciation expense. (2) There is an added restriction
that firms must report compensation. (3) There is an added restriction that firms must have complete data for all variables of
interest, including an interpretable proxy statement.
Variables: FOURCAT is the categorization of the use of measures in bonus compensation into four categories; HCAP, the
composite of three ratios which captures the firm’s use of human capital; EGAL, the ratio of average employee compensation to
average top executive compensation which captures the pay structure; SIZE, the size of the firm, measured as sales; GROW, the
market-to-book ratio.
One limitation of this study is that it uses firms that self-select into a group that discloses labor costs.
The effect of this limitation is to skew the sample to focus primarily on non-manufacturing firms. The
sample of firms is heavily concentrated in financial service firms (54.82%) followed by transportation
and communication firms (20.90%). Due to the uneven weighting of industry classification, sensitivity
tests controlling for industry effects are discussed in Section 4. The sample firms have median sales of
$901.48 million and median employees of 4500.
hazard present in the firm and enable a firm to better extract benefits from its human capital. Thus, the needfor the additional information captured in non-financial measures will not be as great for firms that relyon human capital in an egalitarian environment as opposed to a hierarchical environment. In conclusion,it is argued that the pay structure will moderate the relation between the reliance on human capital andthe use of non-financial measures. This leads to the following hypotheses:H2a. The pay structure moderates the relation between the use of non-financial measures and the relianceon human capital. Specifically, the likelihood of the use of non-financial measures in bonus compensationfor firms that rely on human capital is greater when firms use hierarchical pay structures than when theyuse egalitarian pay structures.H2b. The pay structure moderates the relation between the use of human resource measures and thereliance on human capital. Specifically, the likelihood of the use of human resource measures in bonuscompensation for firms that rely on human capital is greater when firms use hierarchical pay structuresthan when they use egalitarian pay structures.3. Research methods3.1. SampleI limit the sample of firms to those that report in no more than five four-digit SIC codes in the samedivision code.7 Firms are required to have more than 250 employees and report sales, total assets, totalliabilities, common shares outstanding, closing price, the number of employees, and labor costs.8 Thisbrings the sample to 1662 firms prior to the restriction on labor costs. Requiring firms to have labor costsfurther reduces the sample to 312 firms. In addition to the reduction in sample size, this restriction shiftsthe sample from an emphasis on manufacturing firms to one on service-type of firms (e.g., transportation,communication, financial services, etc.). I also require that the firm have an executive bonus programand that the 1997 proxy statement, inclusive of the compensation committee report, be available andinterpretable9 on Compact Disclosure. This results in a final sample of 177 firms. Descriptive data arefound in Table 1.problem. An example of a horizon problem is when managers do not reduce production when demand falls. If managers areevaluated on margin and the production process uses full cost accounting, then a focus on financial incentives may inducemanagers to maintain (or increase) production, thus forsaking the strategic objectives of the organization in order to meetfinancial outcomes. Focusing on non-financial measures, especially human resource measures, directly related to the firm’sstrategy could help mitigate the potential adverse behaviors.7 Highly diversified firms may pursue multiple strategies and rely on multiple resources that differ across business segments,which would add considerable noise to this study.8 Labor and related expenses is data field #A42. Compustat defines this item as “salaries, wages, pension costs, profit sharingand incentive compensation, payroll taxes, and other employee benefits.” According to a representative from Compustat, thisdata field does not articulate with a firm’s income statement; rather, it represents the total amount spent on labor and relatedexpenses regardless of how the cost is accounted for. This is neither a required data item nor is this a required disclosure in thefinancial statements. Therefore, this study uses firms that self-select into a group whose members choose to provide this item intheir public documents. This restriction could have the effect of reducing the variability of the labor intensity ratios since thesefirms probably rely more on labor than do firms that do not disclose labor costs.9 The compensation committee report had to contain enough information to support an accurate coding of the variables.(1) Firms are selected using the Compustat Industrial Files and the Compustat files. Firms report in five or fewer SIC codeswithin one division code. Must report sales, have employees >250 and depreciation expense. (2) There is an added restrictionthat firms must report compensation. (3) There is an added restriction that firms must have complete data for all variables ofinterest, including an interpretable proxy statement.Variables: FOURCAT is the categorization of the use of measures in bonus compensation into four categories; HCAP, thecomposite of three ratios which captures the firm’s use of human capital; EGAL, the ratio of average employee compensation toaverage top executive compensation which captures the pay structure; SIZE, the size of the firm, measured as sales; GROW, themarket-to-book ratio.One limitation of this study is that it uses firms that self-select into a group that discloses labor costs.The effect of this limitation is to skew the sample to focus primarily on non-manufacturing firms. Thesample of firms is heavily concentrated in financial service firms (54.82%) followed by transportationand communication firms (20.90%). Due to the uneven weighting of industry classification, sensitivitytests controlling for industry effects are discussed in Section 4. The sample firms have median sales of$901.48 million and median employees of 4500.
การแปล กรุณารอสักครู่..
