What is the Impact of Minimum Wages on Household Income Distribution?

Concern about the ability of markets to provide income equity for the least able members of the work force has given the minimum wage its strong social appeal. Statutory minimum wages were first introduced in developed countries and later spread to developing countries to control the proliferation of sweatshops in manufacturing industries. The sweatshops employed large numbers of women and young workers, paying them what were considered to be substandard wages. The sweatshop owners were thought to have unfair bargaining power over their workers, and a minimum wage was proposed as a means to make them pay “fairly”.

 

Over time, the focus changed to helping people, especially families, become more self-sufficient. Today, minimum wage laws affect workers in most low-paid fields of employment in over a hundred countries. Supporters of the minimum wage say that it increases the standard of living of workers and reduces poverty. Opponents say that it increases unemployment, particularly among workers with very low productivity due to inexperience or handicap, thereby harming less skilled workers and possibly excluding some groups from the labor market. It therefore fails to reduce poverty. Minimum wage laws have consequently often been judged against the criteria of supporting low-income households and reducing poverty.

 

Support Based on Normative Judgment

 

Although these goals are now widely accepted as proper, there is great disagreement as to whether the minimum wage is effective in attaining them. From the time of their introduction, minimum wage laws have been highly controversial politically and have received much less support from economists than from the general public.

 

Surveys of economists who support the minimum wage have found that their position is based on their personal value judgments rather than economic analysis. These economists believe the minimum wage will promote income redistribution and provide support for low-income households. Nevertheless most of them also believe that raising the minimum wage will lower employment and raise unemployment.

 

An example of these surveys comes from 2007, when Klein and Dompe conducted a non-anonymous poll of supporters of the minimum wage who had signed the “Raise the Minimum Wage” statement published by the Economic Policy Institute. They found that a majority signed on the grounds that it transferred income from employers to workers and resulted in income redistribution. In addition, a majority considered reduced employment to be a moderate potential drawback of the minimum wage hike they supported. The economists were therefore supporting the minimum wage on equity grounds rather than economic efficiency.

 

Various groups have great ideological, political, financial, and emotional investments in issues surrounding minimum wage laws. For example, agencies that administer the laws have a vested interest in showing that “their” laws do not create unemployment, as do labor action groups who derive their political support from members whose jobs are protected by minimum wage laws. On the other side of the issue, low-wage employers affected by the minimum wage have released numerous studies to demonstrate their negative effects on business and employment. The presence of these opposing groups and their vested interests means that the debate on the issue is not always based on dispassionate analysis.

 

But if economic and social policy is going to make sense and be in the interest of the public, it is essential that empirical evidence be brought to examine if the objectives are being met as intended or not. Hong Kong enacted minimum wages in May 2011. Has this made a difference in helping low-income households and reducing income inequality? I report below my attempt to make some preliminary estimates of the effect of minimum wages.

 

Positive Analysis of Minimum Wages

 

To examine this issue I looked for a preliminary answer in the General Household Survey. I examined data in the four quarters before and after the minimum wage became effective, i.e., the periods 2010.Q2 to 2011.Q1 and 2011.Q3 to 2012.Q2. The quarter 2011.Q2 was omitted deliberately as it straddled the date when the minimum wage became effective. Since the sample sizes represent only 0.2% of the population, the accuracy of the estimates may not be precise.

 

My approach is to compare the income distribution of all households with those where there is at least one working member that qualifies for receiving the minimum wage. For minimum wages to be a useful tool for alleviating poverty across households, it is necessary to find these workers to be concentrated among low-income households. If a large proportion of workers that qualify for minimum wages are found among middle- and high-income households then the minimum wage cannot be a useful anti-poverty policy.  

 

Table 1 gives counts of the number of households by 10 decile groups based on household income. I report separately the number of all households and the number of households where at least one working member qualifies for the minimum wage of $28 per hour. The counts are given in cumulative numbers, for example, the lowest decile includes 10% of the households with the lowest incomes, the lowest to 2nd decile includes 20% of the households with the lowest income, and so on.

 

Table 1: Cumulative Number of Households and Households with Minimum Wage Workers by Income Deciles

	Households with  Minimum Wage Workers			All Households
	Cumulative Numbers	Cumulative Share		Cumulative Numbers	Cumulative Share
Four Quarters Before Minimum Wage Introduced (2010.Q2-2011.Q1)
Lowest Decile	14365	0.062	255974		0.110
Lowest to 2nd Decile	55377	0.238	527726		0.227
Lowest to 3rd Decile	78550	0.338	725471		0.312
Lowest to 4th Decile	118209	0.508	964618		0.414
Lowest to 5th Decile	149761	0.644	1188896		0.511
Lowest to 6th Decile	180483	0.776	1415273		0.608
Lowest to 7th Decile	206876	0.889	1660809		0.714
Lowest to 8th Decile	220892	0.950	1885288		0.810
Lowest to 9th Decile	229582	0.987	2120847		0.911
Lowest to Highest Decile	232629	1.000	2327350		1.000
Four Quarters After Minimum Wage Introduced (2011.Q3-2012.Q2)
Lowest Decile	8115	0.067		260462	0.110
Lowest to 2nd Decile	23527	0.195		489762	0.207
Lowest to 3rd Decile	43829	0.362		758652	0.321
Lowest to 4th Decile	61017	0.504		968403	0.410
Lowest to 5th Decile	76934	0.636		1215405	0.514
Lowest to 6th Decile	91618	0.757		1424599	0.603
Lowest to 7th Decile	105844	0.875		1712650	0.725
Lowest to 8th Decile	113938	0.942		1918450	0.812
Lowest to 9th Decile	118291	0.978		2149216	0.909
Lowest to Highest Decile	120953	1.000		2363276	1.000

 

Moreover households with a small number of working members are very likely to have lower incomes than those with more working members, therefore a better (but not ideal) approach is to construct a specific set of income deciles for households that have the same number of working members.

 

I first separated households in these two periods into four groups based on the number of working individuals in the household: 0, 1, 2, and 3+ (denoting 3 or more). Within each and every of these four groups I divided the households into 10 deciles based on their household income.

 

I then added up the number of households in each of the corresponding income deciles to form the total number of households in the income deciles in all four groups of households.

 

In the four quarters before the minimum wage became effective, an estimated 233,000 households (or 10.0% of all households) had at least one member who would benefit from the minimum wage. Some households may have had more than one member who qualified as a minimum wage worker, so this estimate of 230,000 households is still in line with the government estimate of 310,000 workers who would benefit from the minimum wage in May 2011.

 

A Well-known Economic “Myth”

 

In the four quarters following the introduction of the minimum wage, the number of households that qualified as having a minimum wage worker was halved to 121,000 (or 5.1% of all households). This demonstrates the significant success of the minimum wage in helping low pay workers.

 

But a more interesting factor is the distribution of minimum wage workers across households by income deciles. Figures in Table 1 show that in the period before 2011.Q2 households with workers that qualified for the minimum wage constituted only 6.2% of the households in the lowest income decile. This is only half of the total number of households in the lowest income decile who comprised 11.0% of the total. Minimum wage workers do not dominate households belonging to the lowest income decile; rather the former is under-represented among very low-income households. This pattern is repeated again in the period after 2011.Q2.

 

So where do we find minimum wage workers? Among the lowest 20% and 30% of the income distribution in the period after 2011.Q2, only a mere 19.5% and 36.2% of households had minimum wage workers. This was not very different from the distribution of all households by income, which was 20.7% and 32.1% for the lowest to the second and third deciles. Even more significant is that about 51.3% of the households with minimum wage workers were in the four deciles in the middle of household income distribution – the fourth to seventh deciles. Even among the top three deciles, 12.5% of the households had minimum wage workers.

 

One would conclude from these approximate estimates that the middle-income households had a proportionately larger share of minimum wage workers than the low-income households, especially the lowest income ones. It is a mistake to believe that the minimum wage helps low-income households. This has long been a myth well known to labor economists who work on this subject.

 

Provide Basic Income Support

 

The reason for this distribution is that quite a few minimum wage workers may be low-wage earners in middle-income households. Similar findings have been found in empirical analyses in the US, UK and Canada. I previously discussed this in my article “Minimum Wage, Employment and Poverty” (HKEJ, 12 January 2011最低工資、就業與貧窮). The precise reason for this has yet to be investigated in Hong Kong. But what one learns from these empirical results is that the minimum wage will not have a major effect on helping low-income households because the workers it affects are pretty much randomly distributed among the low and middle-income households, and to a lesser extent even among the high-income households.

 

Data from the four quarters after the introduction of the minimum wage show that our results are basically unchanged. So there is no reason to believe that going forward with further increases in the minimum wage will mainly benefit low-income households.

 

It is not difficult to understand why the minimum wage is favored in so many countries. It provides governments with a costless policy (i.e. it does not require government expenditure) with which they can claim they have addressed the problem of poverty. It also provides politicians and labor groups with a convenient instrument to secure votes and clients for membership. But unfortunately it is an ineffective solution.

 

For Hong Kong to genuinely address poverty, it is far more important for members of the government and the Central Policy Unit to study much more comprehensively the real incidence and causes of poverty and not let politics get in the way. Otherwise the Poverty Commission will only be shooting in the dark as the government has already done with its minimum wage legislation.

 

Economists and other political commentators have proposed alternatives to the minimum wage, based firstly on the argument that the minimum wage does not address why there is poverty and fails to remove or alleviate factors and conditions that create many types of poverty. Raising wages merely tackles a symptom not the cause of poverty. Moreover, they argue that the minimum wage is not even the best method to raise the income of low-wage workers.

 

Their suggested alternatives include the likes of a negative income tax and its cousins such as the “guaranteed minimum income” and the “refundable tax credit”. These alternatives have been shown to address the issue of poverty better than a minimum wage, as they would benefit a broader population of low wage earners, do not cause unemployment, and distribute the costs widely rather than concentrating them on the employers of low wage workers.

 

A negative income tax is a system that periodically provides each citizen with a sum of money that is sufficient to live on. Except for citizenship, this basic income is entirely unconditional. There is no means test and the richest as well as the poorest citizens receive it. The basic income is often proposed in the form of a citizen’s dividend (a transfer payment from the government). Proponents argue that the basic income is based on a broad tax base and more economically efficient than the minimum wage, which effectively imposes a high marginal tax on employers causing losses in efficiency.

 

Milton Friedman proposed the negative income tax in Capitalism and Freedom (1962) as means to address the issue of low incomes. The proposal quickly attracted widespread non-partisan support among economists of all persuasions. In 1968 James Tobin, Paul Samuelson, John Kenneth Galbraith and another 1,200 economists signed a document calling for the US Congress to introduce a system of income guarantees and supplements in that year. Both Tobin and Samuelson have also come out against the minimum wage. Winners of the Nobel Prize in Economics who fully support a basic income include Herbert A. Simon, Friedrich Hayek, Robert Solow, Milton Friedman, Jan Tinbergen, James Tobin and James Meade. The provision of a basic income is a far better policy than a minimum wage. This should be considered in Hong Kong.

 

References:

 

Klein, Daniel and Stewart Dompe. “Reasons for Supporting the Minimum Wage: Asking Signatories of the “Raise the Minimum Wage” Statement”, Economic Policy Institute, January 2007, p. 133.