Efficiency Wages and the Economic Effects of the Minimum Wage: Evidence from a Low-Wage Labour Market. Andreas Georgiadis

Transcription

1 Efficiency Wages and the Economic Effects of the Minimum Wage: Evidence from a Low-Wage Labour Market Andreas Georgiadis

2 What we do: Overview -We exploit a natural experiment provided by the 1999 introduction of the UK National Minimum Wage (NMW) to investigate the relationship between wages and supervision and test for efficiency wage considerations in a low-wage sector, the UK residential care homes industry. What we find: -evidence of a wage-supervision trade-off that is consistent with the shirking model of efficiency wages and supports an efficiency wage rationale of why the NMW introduction generated moderate employment effects in the care homes sector.

3 Outline Motivation and Contributions A Simple Theoretical Model The Data and Identification Strategy Results Summary and Conclusions

4 Motivation and Contributions Efficiency Wages (EW) theory offers an explanation of involuntary unemployment. There is a vast number of empirical studies (Krueger and Summers, 1988; Brown and Medoff, 1989) but still the evidence is unpersuasive and inconclusive. The most credible empirical studies (Groshen and Krueger, 1990; Krueger, 1991; Rebitzer, 1995) exploit natural experiments to test the wagesupervision trade-off prediction of the shirking model. Rebitzer (1995) points out that: The difficult econometric problems that hinder the empirical investigation of EW theory make it unlikely that any single study will settle the issue decisively. The empirical fate of EW theory will more likely be determined by evidence from a variety of different investigations each having important limitations and qualifications.

5 Motivation and Contributions We exploit the link between Efficiency Wages and the Minimum Wage to test the wage-supervision trade-off prediction of the shirking model in the UK residential care homes sector The link is based on the fact that: -Efficiency Wages can provide a rationale of the non-negative employment effects of the Minimum Wage (Manning, 1995) -The Minimum Wage provides a natural experiment to test Efficiency Wage theories Our main contributions are: new credible empirical evidence on Efficiency Wages a test of Efficiency Wage models of the Minimum Wage an investigation of a potential explanation of the evidence that although the Care Homes sector was heavily affected by the UK NMW introduction there were only moderate employment effects (Machin et al., 2003).

6 A Simple Theoretical Model Consider an extension of Shapiro-Stiglitz (1984) and Rebitzer and Taylor (1995) that treats supervision as endogenous: -effort is imperfectly observed and the probability of detecting a shirker is given by: P=min{N/L, 1} -where N is the number of supervisors and L is the number of production workers -under the assumed monitoring technology and by firm s profit maximisation the Non Shirking Condition (NSC) is expressed: N e( r + s + q) = L ( w µ e)(1 q) e r s -where is effort, is the time preference rate, is the probability of finding a job, q is the quit rate and the outside option is µ -a Minimum Wage introduction leads to a fall in supervision intensity both in partial and general equilibrium as long as the Minimum wage increases unemployment

7 The Data and Identification Strategy Data collected through the CEP by a survey included the population of care homes conducted around the window of the April 1999 NMW introduction (Machin et al., 2003) Information collected included wage and hours distribution within each home as well as other workers and home characteristics. The survey achieved a reasonable response rate (20%) and based on LFS, the sample of homes that responded was representative of the population

8 The Data and Identification Strategy We are estimating the following system of equations: lns = β + β lnw + β Ψ + u it 1 it 2 i, t 1 0 it lnw = α + α MIN + α Ψ + v it 1 i, t 1 2 i, t 1 0 it (1) (2) -as measures of impact of the NMW, MIN i,t-1 we use the proportion of workers paid below the NMW before the introduction and the wage gap which is defined as follows: GAP i = j h ji max( W j h ji min ji W ji W -implicit identification assumption is that variation in MIN i,t-1 is not driven by variation in initial wages (Machin et al., 2003) ji,0)

9 Table 3: Home Level Wage Effects Preintroduction NMW impact measure Initial period low pay proportion Initial period wage gap Change in log Hourly Wage of non-managers (1) (2) (3) (4) (5) (6) 0.15 *** (0.01) 0.95 **** (0.06) (0.017) 0.86 **** (0.07) 0.13 **** (0.015) 0.94 **** (0.06) (0.017) 0.89 **** (0.08) Controls No No No Yes Yes Yes R No. of observations Initial period low pay proportion Initial period wage Change in log Hourly Wage of non-managers excluding senior carers 0.14 **** (0.01) 0.93 **** (0.06) 0.02 (0.017) 0.86 **** (0.06) 0.14 **** (0.01) 0.94 **** (0.06) (0.018) 0.88 **** (0.08) gap Controls No No No Yes Yes Yes R No. of observations ** statistically significant at the 0.05 level,***at the 0.01 level, robust standard errors in parentheses. Controls include: proportion female, average age, proportion with nursing qualification, proportion of residents who either pay local authority prices for beds or who have their care paid for by the Department of Social Security (DSS) and region dummies. In each specification the change in log hourly wage, instruments and workers controls are computed for the same group of employees.

10 Table 4: OLS versus 2SLS estimates of the wage elasticity of supervision intensity Change in log ratio of managers to non-managers (bodies) Change in log ratio of managers to non-managers (hours) (1) (2) (3) (4) OLS 2SLS OLS 2SLS Change in log average wage (0.38) **** (0.83) (0.41) ** (0.9) Controls Yes Yes Yes Yes R-squared Number of Homes Change in log ratio of managers and senior carers to non-managers (bodies) Change in log ratio of managers and senior carers to non-managers (hours) OLS 2SLS OLS 2SLS **** Change in log average wage (0.48) (0.8) (0.54) (1.01) Controls Yes Yes Yes Yes R-squared Number of Homes **statistically significant at the 0.05 level,***at the 0.01 level, robust standard errors in parentheses. In the 2SLS estimation we use the wage-gap for non-managers and the wage-gap for non-managers excluding senior carers as an instrument for the change in the average hourly wage of non-managers and non-managers excluding senior carers respectively. Controls include: proportion female, average age, proportion with nursing qualification, average intensity of work, proportion of residents who either pay local authority prices for beds or who have their care paid for by the Department of Social Security (DSS), region dummies, response month dummies, whether part of larger organisation and ownership type. In each specification the change in log hourly wage, instruments and workers controls are computed for the same group of employees.

11 Results a 1% increase in the average wage of non-managerial employees reduces the bodies and hours ratio of managers to non-managers by 2.17% and 1.88% respectively a 1% increase in the average wage of non-managerial employees excluding senior carers reduces the bodies ratio of managers and senior carers to non-managers by 2.3% we find evidence that labour demand adjustments counteract the wage-supervision trade-off because of substitution of supervisors for supervisees overall our results are consistent with the shirking model and suggest that lower supervision costs may account for the moderate employment effects of the NMW introduction in the care homes sector

12 Summary and Conclusions We exploit the 1999 UK NMW introduction to test the wage-supervision trade-off prediction of the shirking model and the validity of Efficiency wage models developed to explain the non-negative employment effects of Minimum Wages We find evidence that higher wage growth generated by the NMW introduction caused a fall in monitoring intensity, as measured by different ratios of supervisors to staff Overall our findings provide support to the shirking model of Efficiency Wages and suggest that the NMW may have operated as an Efficiency Wage in the care homes sector