VoxEU Column Labour Markets

Minimum wage effects across heterogeneous markets

Studies often find an aggregated near-zero employment effect when increasing the minimum wage. But the effects might vary if local employers in different regions have different market power. The column examines increases in the minimum wage in Japan and finds a more pronounced negative employment effect in local labour markets where employers had less control over wages.

If we want to estimate the employment effect of the minimum wage, it is important to know the extent of an employer’s labour market power. Undergraduate labour economics textbooks often describe employment effects using a simple competitive labour market model. It describes employers who have little control over wages and immediately reduce employment levels after an increase in the minimum wage. 

But in a monopsonistic labour market, employers have some power to control wages, which means workers are often paid less than the value of the marginal product of labour (MPL). Profit-seeking employers react to an increase in the minimum wage differently from those in a competitive labour market model, and this sometimes leads to a rise in the employment level (Card and Krueger 1994, Manning 2003). Although the two models predict different consequences, research has not given much attention to the role and extent of market power in the employment effects of the minimum wage. In fact, we have little evidence that a monopsonistic labour market exists.

Recently, a growing number of studies have presented evidence of market power (Azar et al. 2017, 2018, Benmelech et al. 2018, Dube et al. 2016, Falch 2010, Ransom and Sims 2010, Staiger et al. 2010). Among them, Azar et al. (2018) highlighted large geographical variations in employer concentration. Analysing a large-scale dataset from Burning Glass Technologies – which collects job vacancy information from approximately 40,000 websites – they found that less-populated commuting zones, and the zones in the Great Plains, tend to have lower employer concentrations. An important factor in the study of a minimum wage is that local labour markets are heterogeneous, meaning employers can respond to an increase in the minimum wage differently across local labour markets.

New evidence

By drawing on Japanese manufacturing census data and administrative wage records, we examined how far an employer was from its competitive optimal decision, and tested whether the employment effect of the minimum wage differed across regions or industries depending on the size of the labour market surplus (Okudaira et al. 2019). We estimated the surplus or wage markdown that employers would face in labour markets with frictions, which we defined as the ratio of the wage rate over the value of the MPL. 

We then estimated the production functions and the elasticities of factor inputs to calculate the surplus in the labour market, and used the estimated surplus to examine whether the effect of the minimum wage on employment differed by the market power of employers in the local labour market. 

Our identification of the minimum wage effect relied on an exogenous policy event. From 2007 onwards, Japanese government policies substantially increased the regional minimum wages over a decade. 

In 2007, the Minimum Wage Act was amended to provide a legal framework to increase regional the minimum wage to, or above, the level of the welfare benefits for each region. The amendment disproportionately affected the prefectures that had initially had a high level of welfare benefits, which were forced to increase their regional minimum wages more than others.

Figure 1 presents a part of the identification variations used in the analysis, that is, the proportion of minimum-wage workers for regions that initially had high welfare benefits (exposed regions) and for non-exposed regions (other regions). 

Figure 1 Proportion of minimum wage workers (waget ≤ mwt+1) by extent of exposed shock

Source: Okudaira et al. (2019), based on data from the Basic Survey of Wage Structures (Japanese Ministry of Health, Labour and Welfare). 

The exposed prefectures experienced disproportionately high increases in the proportion of minimum-wage workers after the amendment compared to the other prefectures. Since this policy set a significantly higher target level for the regional minimum wage before any adjustments by the local authorities, it limited the local authorities’ flexibility in controlling the absolute minimum wage level, alleviating the concern that pre-existing local employment trends may have confounded the results. We also tested for the exogeneity of these events and found that pre-existing local trends had no predictive power. 

Consistent with a standard competitive labour market model, our analysis revealed that plants significantly reduced their employment growth in response to the increases in the minimum wage. Particularly, our baseline estimates suggest that, on average, employment growth in plants in the exposed prefectures was 3.2% to 4.2% lower from 2007 to 2014, compared to those in non-exposed prefectures. But the estimated negative impact masks that an increase in the minimum wage affected the response of plants in our sample differently, depending on the surplus in their local labour markets. 

In response to an increase in the minimum wage, plants that initially experienced a large surplus did not significantly reduce their employment growth. In our preferred specification, the estimated magnitude of the impact was reduced by at least 35%, and even approaches zero for plants with the wage markdown below 0.6 (a relatively large surplus). Therefore, local labour markets are heterogeneous, and the response of a plant to minimum wage shocks depends on the size of the surplus in the labour market that they can exploit.

We are not alone in reporting heterogeneity in the employment effect of the minimum wage. For instance, Harasztosi and Lindner (2019) found heterogeneity in firms’ responses. Job losses were larger among firms in tradable sectors, in which it is more difficult to increase product prices because of foreign competitors. Cengiz et al. (2019) found that the employment effect of the minimum wage is not uniform across US industries, and that disemployment is concentrated in tradable or manufacturing sectors. 

While these two studies look at the heterogeneous effects by considering product market competition, we have focused on labour market competition.1 The employment effect of the minimum wage also differs depending on the employer’s market power in the local labour market. Aggregating the employment effect across local labour markets masks heterogeneity in the employers' responses. This may be a reason that previous studies of a minimum wage have found mixed employment effects.

Policy implications

Our results also suggest that increasing the minimum wage is not a simple redistribution policy. In markets in which employers capture little surplus in the labour market, an increase in the minimum wage is likely to reduce employment growth. Some workers would have to bear the cost of the minimum wage. 

In contrast, in markets where employers can control their wages, increasing the minimum wage can be an efficient policy in boosting the incomes of low earners without a loss in employment. In other words, whether an increase in the minimum wage essentially helps workers or not depends on the frictions in the local labour market. 

Policymakers often take local economic conditions into account when revising the minimum wage, but they pay limited attention to the extent of frictions or market power in the local labour markets. But an increase in the minimum wage can slow down the employment growth of specific industries or regions and so, when local labour markets operate under different market mechanisms, this requires their attention. Our study suggests they should check the frictions in heterogeneous markets before increasing the minimum wage. 


Azar, J A, E Huet-Vaughn, I Marinescu, B Taska, and T von Wachter (2019), "Minimum wage employment effects and labor market concentration", NBER working paper 26101. 

Azar, J A, I Marinescu, M Steinbaum, and B Taska (2018), "Concentration in US labor markets: Evidence from online vacancy data", NBER working paper 24395.

Azar, J A, I Marinescu, and M Steinbaum (2017), "Labor market concentration", NBER working paper 24147.

Benmelech, E, N Bergman, and H Kim (2018), "Strong employers and weak employees: how does employer concentration affect wages?", NBER working paper 24307.

Card, D, and A B Krueger (1994), "Minimum wages and employment: A case study of the fast–food industry in New Jersey and Pennsylvania", American Economic Review 84(4): 772–793.

Cengiz D, A Dube, A Lindner, and B Zipperer (2019), "The effects of minimum wages on low-wage jobs", Quarterly Journal of Economics 134: 1405–1454. 

Dube, A, T W Lester, and M Reich (2016), "Minimum wage shocks, employment flows, and labor market frictions", Journal of Labor Economics 34(3): 663–704.

Falch, T (2010), "The elasticity of labor supply at the establishment level", Journal of Labor Economics 28(2): 237–266.

Harasztosi, P, and A Lindner (2019), "Who pays for the minimum wage?", American Economic Review 109: 2693-2727.

Manning, A (2003), Monopsony in motion: Imperfect competition in labor markets, Princeton University Press.

Okudaira, H, M Takizawa, and K Yamanouchi (2019), "Minimum wage effects across heterogeneous markets", Labour Economics 59: 110–122.

Ransom, M R, and D P Sims (2010), "Estimating the firm’s labor supply curve in a new monopsony framework: Schoolteachers in Missouri", Journal of Labor Economics 28(2): 331–355.

Staiger, D O, J Spetz, and C S Phibbs (2010), " s there monopsony in the labor market? Evidence from a natural experiment", Journal of Labor Economics 28(2): 211–236.


[1] Azar et al. (2019) found that increases in the minimum wage increased employment when the occupational labour market was concentrated, or when the Herfindahl-Hirschman index (HHI) was high. Our study complements their methodology by measuring the firms’ surplus by estimating the production function. This allowed us to capture all sources of friction that the local employers encountered in the labour market.

4,304 Reads