Reviving the Minimum Wage debate (Part 1)

Gordon Lee /

Since the parliamentary “debate” in January this year, which saw Members of Parliament coming out strongly as a united front against implementing a minimum wage, discussions on the pros and cons of a minimum wage seem to have died down.

Government rhetoric asserts that a minimum wage policy has few merits, but significant demerits - such as unemployment, inflation and loss of competitiveness. Lee Hsien Loong was reported to have said, “Singapore has something better than a minimum wage and it is called Workfare.”

This multi-part series will examine these claims from two perspectives: economic theory, and empirical research. In this first instalment, we look at the alleged “unemployment” that is predicted to occur with a minimum wage.

Unemployment in theory

This simple textbook model is easily recognisable by A-Level economics students. Briefly, the downward-sloping demand curve shows that as wages fall along the Y-axis, firms’ demand for workers will increase rightwards along the X-axis. The upward-sloping supply curve shows that as wages increase along the Y-axis, more workers are willing to work for longer, and the supply of labour increases rightwards along the X-axis. The equilibrium level of employment in this perfectly competitive labour market is at E2, with the equilibrium wage at W1.

With a minimum wage set above the equilibrium wage (W1) at the level W min, workers are willing to supply E1 amount of labour, but firms are only willing to employ E3 amount of labour, leading to a fall in employment from E2 to E3 with unemployment as shown. So far, this is what ministers seem to be implying.

There is a problem with this theoretical understanding, as it is based on the assumption that firms operate in a perfectly competitive labour market with practically no ability to set wages. For example, fast food chains and supermarkets seem to have considerable power to determine wages, and the sector can be said to be closer to a monopsony (sole buyer with full control over wages) than a perfectly competitive market (infinite buyers each of small size with no control over wages).


Modelling the monopsonistic market is a little more complex. Although the equilibrium in the free market between Supply (S) and Demand (MRP) for labour is at point C, the monopsonist would choose to employ where his marginal cost (MC) is equal to his marginal benefit (MRP) at point A, hence employing L amount of labour instead of L’ in order to maximise profits. Since a monopsonist has the power to decide wages, he will pay the minimum amount (W) required to secure L amount of labour instead of the equivalent W’ under a perfectly competitive market.

There is also a deadweight welfare loss to society shown by the triangle ACM - which means that because resources are not efficiently allocated, the overall cost to society (in terms of both consumer and producer) is more than the overall benefit to society.

But, if a minimum wage is set at W’, than the monopsonist has no choice but to pay W’ instead of W level of wages, and in order to continue to maximise profits, the monopsonist will increase employment from L to L’. Not only is employment increased, but the deadweight welfare loss is also eliminated with a minimum wage. It’s highly counter-intuitive, but true. In some markets (which are closer to a monopsony rather than a perfectly competitive one), employment increases!

The Singapore economy is a complex one which the first textbook model cannot represent adequately. The economy comprises a mix of competitive and monopsonist markets, and though employment might decrease in more competitive markets, employment increases in more monopsonist markets, mitigating (though not necessarily eliminating) effects of employment/unemployment either way.

Unemployment in practice

“We find that minimum wages ... do not have a negative impact on employment.” [1]

“To sum up, the evidence for the UK presented in this paper indicates that the effect of the introduction of the minimum wage on the probability of employment is insignificantly different from zero for all four demographic groups and in all three datasets used.” [2] (If you are interested in studying the empirical evidence further, reference [2] provides a great overview of research done so far (US, UK, France), and the results.)

Due to the mitigating effects of monopsonist markets, it is no wonder that most empirical studies show little significant overall impact of a minimum wage on employment.

In any case, as past recessions have shown (and the government is quick to point out), any rise in unemployment is partially absorbed by foreign workers. The remaining unemployment has to be weighed against the benefits of a minimum wage, which will be explored in subsequent articles.


While a staggeringly high minimum wage could decrease overall employment, economic theory and empirical statistics demonstrate that a minimum wage set at a modest level* has little (if any) negative effect on unemployment. Unemployment should also be weighed against achieving a country’s other economic and social objectives.

*See http://wrap.warwick.ac.uk/1560/1/WRAP_Stewart_twerp630.pdf (pg. 29) for an elaboration

In subsequent parts of this series, I aim to show that there are strong economic as well as social justifications for the implementation of a national minimum wage.



[1] The Effects of Minimum Wages on Employment: Theory and Evidence from Britain

[2] The Impact of the Introduction of the UK Minimum Wage on the Employment Probabilities of Low Wage Workers