id : pw :
  Intelligencekor
  [representing] intelligence:

Employment shocks from the minimum wage hike and suggestions

First registered: 2018-10-06;     since, no change    
1. Background: Concerns on the employment shocks from the minimum wage hike of Moon's government During the election, then presidential candidate Moon of South Korea promised on the minimum wage of 10 thousand wons per hour (about 8.9 dollars). To keep the promise, Moon's government hiked the minimum wage for this year by 16.4% to 7,530 from 6,470 wons in July in the last year and again for the next year by 10.9% to 8,350 wons. So the hiked minimum wage has been being applied this year, but the responses of employment have been negative so far. The monthly employment increases had been at least at the figure of two hundred thousands in the last year and in January of this year but just wore of the figure of one thousands from February to June and reached the figure of several thousands in July and in August. These sluggish increases of employment lag behind those of working age population so that employment rates (employment over working population) started to drop since February by 0.1 to 0.2 percent point and eventually by 0.3 percent point in July and August. The purpose of the minimum wage is to guarantee a certain level of wage income for workers who get paid as low as around minimum wage. However, concerns on the minimum wage increase is that those workers who get paid around the minimum wage may be low skilled and are likely to lose jobs. Considering that the more the rate of increase in the minimum wage, the larger the chances of job loss. It is likely that Korea may prove such a negative side of the minimum wage increase. The degree of sluggishness in employment growth is so low that it reminds of the employment depression triggered by the 2008-09 world economic crisis. And it is widely believed that the cause of the employment shock of nowadays is the minimum wage hike. However, although the relation between the employment shock and the minimum wage hike looks strong, we cannot exclude the chance that the relation just seems so because of coincidence. Moon's government admits the setbacks of employment but not the relation between the minimum wage hike and the employment shock. In spite of the negative statistics in employment during most of the first half of this year, in July Moon's government decided another minimum wage hike for the next year. Although there are strong sentiments and voices against the minimum wage hikes, Moon's government sticks to them saying that the positive effects will come out over time. The confrontation on the minimum wage hike is not only due to theoretic or ideological differences but also the causal relation between the minimum wage hike and the employment setbacks. Scientific proof of a causal relation may be done either by experiment or by statistical analysis, most social phenomena allow only statistical analysis. However, even statistical analysis cannot fully prove a casual relation but helps to deepen understanding the relation. Although it is just 8 months since the minimum wage hike was implemented and so yet early to do an analysis. But the degree of employment setbacks is so severe that if its cause is the minimum wage then that should be taken into consideration of future decision on minimum wage and also alternatives should be searched for. Thus it is needed to have more quantitative clues on the causal relations. This article attempts to obtain some statistical clues on the causal relations. This article analyzes the employment rates for the past decade. 2. Employment rates and their compositions Employment rate is the proportion of the number of employed people over that of the working population and provided with relevant level statistics at a monthly basis from Economically Active Population Survey (henceforth, EAPS) of KOSTAT (Statistics Korea) which is available at KOSIS (KOrean Statistical Information Service, http://kosis.kr). The latest EAPS covers the figures of employment in August, which are covered in this article. [1] The statistical figures provided in this article is selected part of analyses of some implications regarding impacts of the minimum wage and the whole analysis is provided in the report of power point presentation (PPT) and its underlying EXCEL files: downloadable at PPT and PPT. Some statistical figures mentioned but not illustrated in this article are also in the report. Monthly employment rates have seasonality and so this article track annual averages of monthly employment rates. One kind of annual average is the average of monthly employment rates of January to of December of each year, named 1t12 average. The other kind is the average of monthly employment rates of September of this year to of August of the next year, named 9t8 average. The 9t8 averages are introduced to reflect the latest or August employment statistics. These two averages can be said to be part of the moving averages of the twelve month interval: that is, the moving averages of Decembers and those of Augusts. Moving averages have advantage to smooth out variations and to draw tendencies. For example, if, as in the employment situation of nowadays, monthly employment rates of the last year are strong relative to those of this year, the 9t8 average taken from September of the last year to August of this year mixes good performances of the last year's September to December and bad ones of this year from January to August. And the 9t8 average includes the monthly employment rates of January to of August of this year and so already covers eight months and so will not be very different from the forth coming 1-12 average of this year unless there are unexpected shocks during the last four months of this year; so 9t8 averages have a role of leading indicator. Regarding figures down below, the time point is expressed as, for example, '17. '17 can mean the actual 2017 for the 1t12 average but also can mean the period from the September of 2017 to August of 2018. Since the labor market is the sum of heterogeneous subgroups, it is proper to look into the characteristics of each subgroup: in particular men and women have quite different characteristics in the labor market from the beginning. Korean men's employment rates tend to be above those of the OECD countries' averages while Korean women's tend to be below those of the OECD countries' averages.[2] Probably because of that, the growth of women's employment rates are more distinctive than those of men's. Let alone the different characteristics of employment trend between men and women, sinced the statistics portal KOSIS distinguishes men and women across employer or employment categories, it is better to start with the sex category and keep it across other categories. Figure 1 shows that as for men's employment rate, 1t12 average is smaller than 9t8 average in '17. It is when 9t8 averages are smaller than 1t12 averages that the 1t12 average tends to decline. Men's employment rates are rather volatile in that they have periods of decline for '14-115 let alone '07-'09 overlapped with the global financial crisis. Men's employment rates are already higher than the OECD countries' average of them, which may bring about such a volatility. So it is hard to tell whether the 9t8 average's smallness in '17 relative to the 1t12 average may reflect the influence of the minimum wage hike or may be part of variations. On the other hand, women's employment rates have shown a steady rise since '09 and the 9t8 average for women is larger than the 1t12 average. But in '17, the 9t8 average for women is very close to the 1t12 average and so there is a chance of reversal between the two in position. Thus if we can find cases of such a reversal among some subgroups of employment for women, that may be indicative of the influence of the minimum wage. Below. some employment subgroups for women that are dense in non-permanent jobs and so are likely to be affected by the minimum wage will be highlighted. Figure 1. Employment rates of men and of women Employment rates of men and of women Sources: KOSTAT, Economically Active Population Survey (henceforth, EAPS); processing is done by the author. As for age groups, jobs for the age group of 15-24 year old and that of 60 or more year old women's are likely to be and are non-permanent.[3] Figure 2 shows that the 9t8 average of the age group of 15-24 crawls down its 1t12 average in '17 which implies a strong chance of a upcoming drop of the 1t12 average. The 9t8 averages of the age group of 60 or more year old women has continued to be larger than its 1t12 average and so 1t12 has continued to rise too since '10. But the 9t8 averages of the age group of 60 or more year old get very close to its 1t12 average in '17, which hints on a chance of the 1t12 average getting smaller in '18 (if the employment rates after August of this year get worse). Indeed, the 9t8 average of the age group of 60-64 year old women is already smaller than its 1t12 average in '16 (as in the companion report). So, among women that show a steady rise of employment rates is also, a vulnerable, there are subgroup that are vulnerable to minimum wage shocks. Figure 2. Employment rates of women by age: for 15-24 and for 60 or more Employment rates of women by age: for 15-24 and for 60 or more Sources: KOSTAT, Economically Active Population Survey (henceforth, EAPS); processing is done by the author. The proportions of non-permanent jobs are highest for the group of middle school graduation or lower education attainment and second highest for high school graduation. As for the group of women high school graduation, its 9t8 average drops below its 1t12 in '17. On the other hand, as for the group of middle school graduation or lower education attainment, its 9t8 average has continued to decline and to be under its 1t12 average, but its 9t8 average has got close to its 1t12 average in '17. As for the group of high school graduation, men also show a similar movement of employment rates in '17 (with a larger extent), Contrarily, the group of men of middle school graduation or lower education attainment shows a similar trend with that of women, which implies a different employment characteristics from the rest. A hypothetical explanation for this characteristics is that, given the fact that older generation has lower education attainment, the people of middle school graduation or lower education attainment gets dropping out of employment pool as they get old. This hypothesis cannot be verified from the EAPS because the EAPS does not provide a cross tabulation of statistics both by age and by education level. Figure 3. Employment rates of women by education level: for middle school graduation or lower education attainment and for high school graduates Employment rates of women by education level: for middle school graduation or lower education attainment and for high school graduates Sources: KOSTAT, Economically Active Population Survey (henceforth, EAPS); processing is done by the author. Employment rates by age and education level, discussed above are based on the population characteristics. On the other hand, the grouping of employment now discussed is based on the job characteristics, and so there cannot be a corresponding grouping of working age population. So the decomposition of employment rates for women, men or the whole will be made; that is, keeping each denominator of an employment rate each of men, women or the whole, the nominator, or the number of employment is decomposed by job category. These indicators has an advantage to the level variables of employment in the sense that they reflect variations of working age population. Jab status, now addressed as job characteristics, classifies jobs first into employees on payrolls and the rest. Employees on payrolls are further classified into regular, temporary, daily workers,[4] and the others not on payrolls are into self-employed and unpaid family members. The author will name the job of temporary and daily workers in combination irregular job (in contrast with regular job). The self-employed will be the sum of those who employ any other than oneself and those who do not employ any other but oneself. As for the minimum wage shock, those of irregular jobs and those of self-employed are in concern. The former show higher share of non-permanent jobs than those of regular job and the latter is supposed to be hit because they mostly have to pay for the hiked minimum wage. Figure 4 shows that 9t8 average of those of irregular job start to drop down at '15 (most of irregular jobs are temporary ones, though both those of temporary job and those of daily job have a similar trend as in the companion report). On the other hand, the 9t8 average for self-employed women bounced back in '16 but showed a slowed down and a close-up to the 1t12 average in '17 hinting on a chance of upcoming decline of 1t12 average. Though not illustrated in this article but in the companion report, the 9t8 average for those self-employed without employment is already smaller than the 1t12 average for them in '17 and accordingly the 9t8 average for those self-employed with employment is still larger than the 1t12 average for them. This contradiction can be explained either by the fact that the self-employed with employment has larger capital and better business shape than the self-employed without employment or by that fact that as the government subsidizes those employers who are likely to pay around minimum wages, the self-employed who report no employment but has unofficial employment turns out to report their employment. Figure 4. Employment rates of women by job status: for those of irregular job and for self-employed  Employment rates of women by job status: for those of irregular job and for self-employed Sources: KOSTAT, Economically Active Population Survey (henceforth, EAPS); processing is done by the author. As for task characteristics, sales and menial labor are two tasks that are high in non-permanent job's share. Figure 5 shows that the 9t8 average for sales women started to drop a little in '16 but further dropped in '17, implying the impact of the minimum wage hike. The 9t8 average for women of menial labor started to drop apparently in '16 and dropped again '17 but with a lesser extent, implying that they experienced employment shock already in '16 but further shocks in '17. The year 2016 faced manufacturing employment shocks (particularly in shipping industry; women also experienced job shocks in manufacturing industry), and since March of 2017, some sectors, including the retail sector in focus, have been hurt by the retaliation of China on the US's deployment of THHAD (Terminal High Altitude Area Defense) on South Korea [5], and in 2017 the minimum wage hike set in. These shocks in a row seem to have impacts on the employment rate for women of sales and menial labor in '16 and '17. Figure 5. Employment rates of women by job charactersitics: for sales and for menial labor  Employment rates of women by job charactersitics: for sales and for menial labor Sources: KOSTAT, Economically Active Population Survey (henceforth, EAPS); processing is done by the author. Note*: The portion of each component of the nominator of the female employment rate by job status to the denominator of the number of the female employment rate. As for sectors, both the wholesale/retail sector and the lodging/dining sector where shares in non-permanent job are high show the decline of the 9t8 average in '17. The wholesale/retail sector leveled in '16 but dropped in '17 for the 9t8 average. For the lodging/dining sector, the 9t8 average was slightly above the 1t12 average in '16 but the 9t8 average got down below 1t12 average in '17. The two sectors were already weakened in employment in '16 but were even more seriously hurt in '17, hinting on the minimum wage shocks. Figure 6. Employment rates of women by sector: for wholesale/retail and for lodging/dining Employment rates of women by sector: for wholesale/retail and for lodging/dining Sources: KOSTAT, Economically Active Population Survey (henceforth, EAPS); processing is done by the author. Note*: The portion of each component of the nominator of the female employment rate by sector to the denominator of the number of the female employment rate. 3. Implications and suggestions Since the minimum wage hike and employment shocks happen simultaneously and increased wages are likely to lead to employment suppression, it may be believed that the ongoing employment shocks are due to the minimum wage hike. Although the employment rate of women steadily has increased for the past decade, the recent data reflecting the months of this year hints on a chance of setbacks of employments. And some subgroups of women that have high shares of non-permanent jobs show a sudden drop of the annual average of employment rates that cover the months of this year so far, which hints strongly on a negative impact of the minimum wage hike. Such subgroups are five: the 15-24 age group, the group of high school graduates, the group of sales task, and the group for the wholesale/retail sector and the group of lodging/dining sector..[***] There are two groups that imply not only the minimum wage hike of this year but also other shocks in the previous years: the group of irregular jobs and the group of menial labor. There are groups too that do not seem to be affected by external shocks but keep a decreasing or increasing trend in employment rates: the age group of 60 or more (decreasing) and the group of middle school graduation or lower education attainment (increasing) and the group of the self-employed (increasing). The group of middle school graduation or lower education attainment seems to reflect on the demographic change in education attainment (people of lower education attainment get older and are likely to drop out of the employment pool). The age group of 60 or more tended to seek employment, but there is a chance of reversal of the trend as in the slow-down in employment rate growth as in the annual average of employment rates covering the months of this year and the age group of 60-64 already shows such a reversal. As summarized above, among the ten groups that may be vulnerable to the wage shock, six are suggestive of the wage shock, two are suggestive of a mixture of different shocks and the rest two. So this shows that a seemingly causal relation between two simultaneous instances cannot be unanimously proved to be casual even through detailed analysis of statistics: some results strongly hinting on casual relation but others looking indifferent or even opposite to expectation based on the casual relation. the negative employment signal raised by the minimum wage is strong enough already. If there are voices of concerns regarding a policy and statistics or statistical analyses are corresponding to the voices to any extent, then authorities responsible for the policy should take the concerns seriously. However, the difficulty of proving casual relation does not justify incapability in practice. Although Moon's government asks people to wait, the minimum wage is a desperate issue to those who have to pay for the wage or to those who may lose jobs. It undermines the Raison d'être of any policy if the policy neglects impending pains due to the policy and pursues only a long term goal. It is not to mention that if the people affected by a policy are in rather economically weak position, then the negative side of a policy should not be overlooked, If the purpose of the minimum wages is to enhance a living standard of people. Let alone the negative effect of the minimum wage, the minimum wage is not the only policy instrument to increase incomes of people of lower earning. It is an inherent nature that the stronger the intervention on the market that balances out demand and supply in terms of price such as wage, the more probable distortion of the market. Thus there are already suggested and implemented policy instruments that minimizes the interruption on the market. For example, Earned Income Tax Credit (shortly, EITC) compensates for a certain portion of the difference between a designated income and the actual wage income at the stage of taxing (so EITC called also negative tax). EITC is is implemented in South Korea too albeit a small scale in the size of tax credit: on average hundred thousand wons in a year (hundreds US dollars). Moon's government is also positive to EITC and extends the budget of 3.8 trillion wons (3.4 billion US dollars) and the number of beneficiaries to three million households; yet the average credit per household is a little more than a million wons (a thosand dollars). Moon's government is subsidizing about 3 trillion wons (about 2.7 billion US dollars) for the employers who are likely to be hurt by the minium wage hike, EITC can be further extended by including such subsidies. Basic Income, which tend to have more cases of experimentation or limited implementation nowadays than in the past, is a strong alternative that does not intervene on the market price but guarantees a certain level of income. Basic income is even simpler than EITC in that basic income is distributed to anyone without means testing. Just considering that if basic income has to cover a level of minimum living standard, then the size of budget has to be too large, So beginning with a low level of basic income or limiting its coverage to a group of a certain demographic profile, we can help to build consensus on basic income. Child care allowance, being practiced in Korea, is a kind of basic income for infants and toddlers that is taken by their parents on behalf of them. The 500 thousand won (450 US dollars) allowances for the age group of 19-24 of the City of Seongnam (sharing a border with Seoul from South) is another example of basic income.[5] Once the consensus gets stronger, then the coverage and the benefit of basic income can be enlarged. As the minimum wage has increased so fast, the decision process calls for reform. (1) Above all, in spite of such a huge impact of the minimum wage, the decision for the minimum wage is not based on a wide consensus. The minimum wage practiced nationwide should be decided not by the government but by the national assembly which is supposed to channel voices of people in its decisions. (2) Along with the nationwide minimum wage, if there is a regional need to have a higher minimum wage than the national one, then a regional level minimum wage should be decided. Regional level minimum wage would be decided by the provincial or local governments considering the economic dimension of the minimum wage. Currently, there are some local governments that designate 'living wage' and use it as a guide in paying for the employed by the local government or agents under the direction of the local government. For regional minimum income, they only have to apply such as living wage region-wide. (3) Lastly, as the minimum wage increased to such a higher level, the need for restructuring fringe benefits has become stronger. The National Assembly extended the coverage of the minimum wage to some fringe benefits (that is, fringe benefits are also counted as hourly wages), which helped to rationalize wage structures. However, the weekly holiday allowances remain to be practiced (A part time worker on a payroll of of more than 15 hours a week should be paid for the equivalent of a 8 hour work day as a fringe benefit). Because of that, some employers avoid the payroll of more than 15 hours and part time employees willing to work more than 15 hours have to seek for another employers, which is a dead loss to both sides. Such peculiar features of fringe benefits of South Korea was admitted as a compensation for low hourly wages, but, as the minimum wages increase, the irrationality of them have surfaced. While inequalities faced by part time labor in Korea should be corrected and part time labor should be encouraged as in Netherlands where a huger portion of women are part timers,[6] the distorted structure of wage and fringe benefits should be reformed. ⚹: This article's Korean version was sponsored by Korea Social Science Institute (KSSI, http://kssi.jinbo.net/) and published in its quarterly journal 'Trends and Prospects' (Korean)(Fall, 2018, no. 104). KIM Seokhyeon, partner of Intelligencekor, has written this article in representation of the company. The Korean version of this article is also available: Korean version. ⚹⚹: Companion report (in Korean) download: PPT report link; EXCEL files link. ***: The published article in Korean counted the six subgroups, not five, of implying the minimum wage shock by mistakenly including the group of the self-employed women 1: Employment statistics are provided every month in Economically Active Population Survey (henceforth, EAPS) by KOSTAT. Economically Active Population covers the people of 15 or older except the institutionalized population (people in military or alternative service and prisoners). People in employment is defined as such: (1) people who worked for more than an hour in the surveyed week; (2) people who are unpaid but work for more than 18 hours in a week for family business; (3) people of temporary off duty because of illness, vacation, bad weather, labor strike and so on. 2: the definition of working age population differs between Korea and the OECD: OECD sets the upper limit of age at 65 while Korea does not. The statistics of employment rates of the OECD countries is provided at: data.oecd.org/emp/employment-rate.htm. 3: KOSTAT also provides statistics for job types or status that can infer job quality as companion statistics to the EAPS across many subgroups of employment whose official name of the survey is Additional Survey by Employment Type(shortly, ASET). According to the ASET, the two age subgroups of 15-19 and of 60 and more have more non-permanent jobs than any other age subgroups. The statistics does not distinguish the age subgroup of 20-24 but just the whole 20's. But given the high rates of college attendance of the age subgroup of 20-24, most of the age subgroup of 20-24 cannot have permanent jobs. 4 Regular jobs in Korea are broader than permanent jobs. Permanent jobs in Korea are defined as jobs with contracts without any defined period of employment and full fringe benefits. In addition to permanent jobs, regular jobs include jobs with contract with a definite time period, though repeatable, and short of full fringe benefits. 5: At the summit between the two countries in the last November China promised on the drop of retaliation but South Korean side has not seen yet the promised recovery of the relation) 6: To help local merchants, the City of Seongnam pays out coupons, not cash, that can be used for purchase in the local or traditional markets. Though this approach is different from that of basic income helped to garner the support of the local merchants for such a redistribution. 7: According to an article of the Economist under heading Why so many Dutch people work part time(May 12, 2015), Netherlands legislated the rights of part time employment and had quality part time jobs and so part time labor is not regarded as discrimination but as a matter of choice. In particular, the 80 percent of the women's jobs are of part time. Link: https://www.economist.com/the-economist-explains/2015/05/11/why-so-many-dutch-people-work-part-time


Notes