Monday, March 21, 2022

Minimum Wage Revised Part I: Theoretical Considerations

So the government has finally announced a new revision to Malaysia's minimum wage, two years after the last one. This time though, it's a whopping 25% increase to RM1,500, from the RM1,200 in 2020. Even after all these years (nine to be exact), the minimum wage continues to be the subject of a lot of arguments, so I thought I'd lay out some of the theory and Malaysian evidence (such as it is).

First of all, what we learned in Econ 101 is that when you establish a price above that of the market determined price, quantity supplied increases while quantity demanded decreases, and the market does not clear. In the context of labour, this implies higher unemployment, as more people are willing to work, but less employers can afford to take them on. But in empirical studies, this generally does not happen with the minimum wage. Why?

Fundamentally, critics of the minimum wage suffer from the lump of labour fallacy, which is a subset of zero-sum thinking. This is the assumption that there are a fixed number of jobs, and that the partial equilibrium approach of Econ 101 is the correct one. The preponderance of evidence that minimum wages do not in fact create higher unemployment suggests that something else is going on. The favourite hypothesis explaining this is that minimum wages also create higher demand for goods and services, given that low income individuals also have a high marginal propensity to consume. Higher minimum wage leads to higher demand for goods and services leads to higher demand for labour, which then neatly cancels the unemployment effect of an artificially higher wage rate.

Real life is of course not quite so neat. My personal belief is that there are both employment and disemployment effects, and the final outcome would depend on the elasticities of labour demand and supply, the consumption basket of low income workers, and the goods and services provided by industries which employ them. The real answer, as in almost all questions in economics, is, "it depends."

I also think that the distribution of wages matters as well. There's a reason why not all minimum wage studies show benign effects. When a minimum wage rises too fast, or covers a bigger portion of the wage structure, that's when I think we start to see the disemployment effects start to predominate. Where that threshold lies is really an empirical question (eg like this).

Moving on, a second criticism of a minimum wage is that it will cause prices to rise, or inflation to accelerate. This one's a lot easier to deal with:

A price level increase is exactly what is supposed to happen. To be more precise, we are looking not for inflation, but a very specific type of inflation.

What are the main price level differences between high income economies and low income economies? One, prices of goods are cheap relative to income in high income economies and two, prices of services are roughly the same (or be even more expensive) relative to income. Oh, and asset prices (ie housing) also becomes more expensive, but I'll leave that aside for now. So if the goal is to create a high income economy, a faster rate of inflation in services relative to goods is a necessary symptom.

Insofar as most goods have some labour component, there will be an increase in prices wherever low cost labour is used. Having said that, it's not the type of inflation that should be feared. Labour compensation in Malaysia is just 37.2% of GDP (2020). That implies that a 10% increase in the wage bill would only increase prices by 3.7%, and workers would generally benefit from higher real incomes.

This is obviously a gross oversimplification and any estimate of costs and benefits, and as in the discussion of employment above, would depend on multiple parameters such as patterns of consumption, the market power of firms to set prices, and changes in the overall wage structure. For example, those in the middle to upper half of the income distribution would I think lose out, though the impact decreases as we go up (due to lower share of consumption relative to income). Nevertheless, I would generally plop on the side that the benefits outweigh the costs to society as a whole.

Third on the theoretical aspects of a minimum wage, is labour productivity. Critics contend that wages should mirror productivity and any increase in wages can only be accommodated by higher productivity. This is bogus, because:

Labour productivity in economics terminology does not mean the productivity of labour. It is calculated based on the amount of valued added created in firms, divided by the labour force. But value added is not just a function of how hard your labour force works, it is also a function of pricing power, industry structures, corporate strategy, capital intensity, and the like. It has as much to do with marketing as it does with hours worked. As bizarre as it might sound, raising prices will also raise "productivity." Apple employees are are highly paid and productive not because they are super smart and work hard (though that doesn't hurt), but because Apple has created a brand and an ecosystem that allows it charge super-normal prices. Labour productivity is as much the responsibility of management as it is of workers.

Last, is the fear that SMEs would not be able to handle the higher wage costs. This is true...sort of. Based on the 2015 census of SMEs, most small and medium businesses should be able to handle the increase with little difficulty. Their wages already averaged RM2,180 and RM2,532 respectively even back then. While there are particular sectors which are close to the line (notably in agriculture), and bearing in mind the differences between average and median, it's not too much of a stretch.

Micro industries are another matter, with an average of RM1,096 in 2015. Even taking into account salary increases since 2015 (2018 was a particularly good year for labour compensation), they'd probably average around RM1,400 about now. Hence the importance of the exclusion of micro industries (<5 employees) from this new revision, since the average for micro industries is about 3 employees. Having said that, this kind of policy threshold could potentially drive perverse and/or avoidance behaviour, never mind the difficulty in enforcement. Also the welfare loss of potentially 2 million workers not being covered by minimum wage legislation is nothing to ignore. We've all heard plenty of anecdotal evidence of workers being underpaid, or forced to take on contractual relationships that subvert the spirit of the Employment Act. A topic for another day.

Next up, examining the impact on wages.

No comments:

Post a Comment