Levantamiento y captura de la información

Many students are initially surprised to see the relative attention paid to labour supply by economists. Perhaps based on their own experience, their caricature of a labour supply deci- sion is that virtually everyone wishes a full-time job or career. However, is it really the case

LO1 LO1

ben40208_ch02_031-073.indd 32

33

CHAPTER 2: Labour Supply: Individual Attachment to the Labour Market

that most people participate in the labour market, working the standard workweek of around 40 hours? As we shall see, there is actually considerable variation in the degree of attachment of individuals to the labour market.

Labour Force Participation

The labour force participation decision is basically a decision to participate in paid labour market activities as opposed to other activities such as unpaid work in the home, volunteer work, education, or retirement. As such, it influences the size and composition of our labour force and has an impact on household activities, education, and retirement programs.

The policy implications of these changes can be dramatic. Changes in the size and com- position of our labour force affect our growth and unemployment rates, as well as the occu- pational and sex composition of the labour force. The latter, in turn, affect such factors as relative wages, demands for unionization, daycare, and equal pay and equal employment opportunity legislation. Changes in household activities can involve family formation and mobility. Retirement programs can be affected insofar as new labour force participants will add contributions to pension funds, while those who retire (i.e., do not participate in the mea- sured labour force) will be a drain on the funds.

As illustrated in Figure 2.1 , the labour force consists of those persons in the eligible popu- lation who participate in labour market activities, as either employed or unemployed. The eligible population is that portion of the population that is surveyed as potential labour force

FIGURE 2.1 Labour Force Concepts (numbers refer to January 2011)

The labour force is the sum of those individuals either working (employed) or not working but search- ing for work (unemployed). The unemployment rate is the percentage of the labour force that is unemployed, while the labour force partici- pation rate is the percentage of the eligible population in the labour force.

NOTES:

1. Population figures in millions .

2. Labour force participation rate: LFPR 5 LF/POP 5 18.66/27.84 5 67.0% . 3. Unemployment rate: UR 5 U/LF 5 1.45/18.66 5 7.8% .

SOURCE: Adapted from the Statistics Canada publication, “Labour Force Information,” Catalogue 71-001-X, February 2011, http://www.statcan.gc.ca/pub/71-001-x/71-001-x2011001-eng.pdf . Accessed February 15, 2011.

Population of Canada

Eligible Population or Potential Labour Force Participants (POP)(millions) (civilian, non-institutional population 15 years and older, excluding Yukon, Northwest Territories, Nunavut, and those living on Indian reserves)

27.84

Labour Force (LF): 18.66 Not in Labour Force (NLF): 9.18 (working or actively seeking work) • Students (not working)

• Retired persons • Household workers

• “Discouraged workers” and others not searching for work

Employed (E): 17.21 Unemployed (U): 1.45

(working) (not working, but searching for work)

ben40208_ch02_031-073.indd 33

34 PART 1: Labour Supply

participants (i.e., civilian noninstitutional population, 15 years and over, excluding Yukon, Northwest Territories, Nunavut, and those living on Indian reserves). Persons from that potential population of labour force participants (POP) are categorized as either in the labour force (LF) or not in the labour force (NLF). Those in the labour force are either employed (E), or unemployed (U), with the latter being not employed but seeking work. People are classi- fied as “employed” if they did any work at all in the survey period (even one hour). They are also considered as employed if they are normally employed but happen not to be at work at the time of the survey because, for example, they are ill or on strike. Note the definition of “work” is left somewhat vague, but it is designed to capture those working for pay, either in the formal labour market or implicitly in self-employment. Students may work hard on their courses, but this is not considered being “employed.” Only students who work for pay are considered employed. Accordingly, those not in the labour force are usually students, retired people, persons in the household, those unable to work, or some “discouraged” workers who have simply given up looking for work. According to labour force definitions, the latter are not categorized as unemployed because they are not seeking work.

The labour force participation rate (LFPR) is the fraction of the eligible population that participates in the labour force (LFPR  5  LF/POP). The unemployment rate (UR) is the pro- portion of the labour force that is unemployed (UR  5  U/LF).

The Canadian Labour Force Survey (LFS) conducted by Statistics Canada is currently based on a monthly sample of approximately 54,000 households, yielding detailed labour force data on approximately 100,000 individuals. The results are published monthly in “Labour Force Information” (Statistics Canada, Catalogue No. 71-001XWE), with the survey described in more detail in that publication.

While the LFS provides the most frequently used estimates for our labour force, employ- ment, and unemployment figures, other sources are available. In particular, the Canadian census is conducted every five years, the most recent being in 2006, referring to activity in 2005. The census is more comprehensive (not being based on a sample from a larger popula- tion), and consequently includes richer details on such factors as employment and unemploy- ment by industry and occupation. However, its use is limited because it is conducted only every five years, there is a considerable lag before the results are published, and its reliability on labour force issues may be questioned because, unlike the Labour Force Survey, it does not focus only on labour force activity.

Labour force participation rates have changed significantly over the 20th century, espe- cially for women. Figure 2.2 shows trends in participation based on census data. As we can see, male participation has declined from just over 90 percent of the eligible population to 73 percent by 2000. Women’s rates were below 30 percent as recently as 1961, but they increased by 30 percentage points over 40 years to 61 percent in 2001. What factors might account for these changes? Was it primarily improvements in contraception or changing atti- tudes toward women’s work that led to increased female participation? Did women’s wages rise since 1960, inducing more women to enter the labour force? Did their husbands’ wages fall, requiring more women to work to support their families? A well-developed economic model would aid in analyzing these explanations and also provide alternative, testable hypotheses.

Figure 2.2 hides as much variation in participation as it shows. For example, participation varies significantly by age and marital status. We focus on these demographic and lifecycle dimensions of labour supply in Chapter 4. Table 2.1 highlights another important aspect of cross-sectional variation in participation: differences across countries. For this sample of countries, labour force participation ranges from a low of 44.9 percent in Egypt to 77.6 per- cent in Denmark. The Canadian rate is at the top end of the range. A careful examination of the numbers in Table 2.1 yields at least two insights. First, most of the variation in overall participation is due to differences in the participation rate of women. Second, participation

www.statcan.ca/english/ census96/list.htm www.statcan.ca/english/ Subjects/Labour/LFS/ lfs-en.htm ben40208_ch02_031-073.indd 34 ben40208_ch02_031-073.indd 34 9/8/11 10:18 PM9/8/11 10:18 PM

35

CHAPTER 2: Labour Supply: Individual Attachment to the Labour Market

10 20 30 40 50 60 70 80 90 100 Men Women 1991 2001 1981 1971 1961 1951 1941 1931 1921 1911 1901 P articipation rate Year

rates for women seem to be positively related to the level of economic development, with the richer countries having higher female participation rates, or at least smaller gaps between men and women.

This second point is illustrated in Figure 2.3 , where we plot the participation rates against one measure of economic development, the logarithm of per capita GDP (national income). One guiding question—which can be only suggestively addressed, given the exploratory nature of this exercise—is whether the labour force participation patterns we observe represent different points along an upward-sloping labour supply function. While it is reasonable to assume that wages for both men and women are higher in richer countries, there are also many other fac- tors that differ across countries. That said, what might we potentially learn about labour supply behaviour from these figures? The top panel shows that male participation is generally declin- ing with per capita income, and thus possibly with higher wages. Does this mean labour supply for men slopes downward? Not necessarily. Perhaps as countries get richer men do not have to work as hard or for as long in their lives? Perhaps they are better able to afford early retirement? This may even be easier if their wives are able to work at higher-paying jobs.

In the bottom panel, we see a strong upward-sloping relationship between national income and female participation. Again, this is far from definitive evidence, but the data are consistent with richer countries having higher-paying opportunities for women (i.e., higher wages) and thus higher female participation rates. Possibly, in richer countries women’s fertility is also

FIGURE 2.2 Labour Force Participation Rates by Sex, Canada, Census Years 1901–2001

Labour force participation rates for men and women are plotted by census years, beginning in 1901. The general trends suggest a gradual reduction of labour force attachment for men, in contrast to a sharp increase for women, beginning in the 1950s.

SOURCES: M.Gunderson, “Work patterns,” in Opportunity for Choice: A Goal for Women in Canada, ed. G. Cook (Ottawa: Statistics Canada, 1976), p.97 for 1901–1971, reproduced by permission of the Minister of Supply and Services Canada. Figures for 1981 were computed from the 1981 Census, Catalogue 92-915, Vol.1, Table 1, p.1-1.Figures for 1991 were computed from the 1991 Census, Catalogue 93-324, The Nation: Labour Force Activity, Table 1. Figures for 2001 were computed from the 2001 Census Catalogue a550490XCB2001001, www12.statcan.ca/english/census01/products/ standard/profiles/RetrieveProfile.cfm?Temporal 5 2001&PID 5 56178& APATH 5 1&RL 5 6&IPS 5 95F0490XCB2001001 .

ben40208_ch02_031-073.indd 35

36 PART 1: Labour Supply

TABLE 2.1 Labour Force Participation Rates by Sex, Various Countries, 2003

Country Abbreviation Overall Male Female Difference

Denmarka _{DNK 77.6 82.2 73.0 9.2}

Bangladeshb _{BGD 72.1 87.2 55.9 31.3}

Canadad _{CAN 67.5 73.6 61.6 12.0}

Philippines PHL 66.2 82.3 52.8 29.5

United Statesd,e _{USA 64.3 73.5 59.5 14.0}

Russiaf _{RUS 63.3 70.0 59.2 10.8}

Australia AUS 62.7 71.0 55.9 15.1

United Kingdome _{GBR 61.9 70.6 55.3 15.2}

Jamaicag _{JAM 61.4 71.4 53.0 18.4}

Hong Kongd _{HKG 61.4 72.0 51.6 20.4}

South Koread _{KOR 58.4 74.6 48.9 25.7}

Mexico MEX 57.8 81.1 38.1 43.0

Argentinah _{ARG 57.1 71.5 45.8 25.7}

Germany GER 55.6 65.4 49.3 16.1

France FRA 54.8 62.6 49.2 13.4

Polandi _{POL 54.8 62.2 48.0 14.2}

South Africa ZAF 54.1 61.2 47.8 13.4

Pakistanj,d _{PAK 50.5 82.7 16.2 66.5}

Italy ITA 49.2 62.3 37.1 25.2

Egyptk,d _{EGY 44.9 68.6 20.2 48.4}

NOTES: Participation rates for the population 15 years and older, unless otherwise noted. “Difference” is the difference between the male and female participation rates. The list of countries is ranked in descending order, from the highest to the lowest overall participation rate.

a. Ages 15–69 g. Ages 141

b. 1999–2000 h. Urban population only c. Excludes indigenous population living on reserves i. Excluding conscripts d. Excludes armed forces j. 2001–2002

e. Ages 161 k. 2001

f. Ages 15–72

SOURCE: Copyright © International Labour Organizationa (ILO Department of Statistics, http://laborsta.ilo.org/). For detailed explanation and definitions, see the related discussion on the ILO Web site, plus ILO Yearbook of Statistics, 2004 (Table 1A)

lower, which further facilitates their movement into the labour force. Notice that there are other variables besides income that explain differences in participation rates. Countries above the regression line have relatively high rates of female participation for their level of devel- opment, while countries below the line have higher rates. Among lower income countries, Bangladesh, the Philippines, and Jamaica have higher female participation rates, especially compared to Pakistan and Egypt. Perhaps differences in culture or attitudes to women’s mar- ket work explain these differences. A good model of labour supply should be able to account for such differences. Taken together, these results point to a positive link between economic development and a reduced gap in the participation rates between men and women. As countries get richer and as women’s education and labour market opportunities improve, a

ben40208_ch02_031-073.indd 36

37

CHAPTER 2: Labour Supply: Individual Attachment to the Labour Market

FIGURE 2.3 Male and Female Participation Rates by Level of Economic Development

This figure shows how male and female participa- tion varies with the level of economic development as measured by per capita GDP. In addition to the data points indicated by the coun- try abbreviations (explained in Table 2.1 ), we also show the estimated regression line relating the participation rate to log per capita income. The top panel shows that the male participation rate is generally declining with national income, while the second panel shows that women’s participation rises with national income.

NOTES:

1. This figure plots the participation rates for men and women against the per capita national income of the country (in natural logarithms). The participation data and country abbreviations are detailed in Table 2.1 . Per capita GDP is taken from the Penn World Tables (2002). This “PPP” (Purchasing Power Parity Adjusted) measure of per capita GDP accounts for differences in the cost of living between countries. We also show the estimated OLS regression line for each series as a function of log per capita income.

2. The GDP data are taken from Alan Heston, Robert Summers, and Bettina Aten, Penn World Table Version 6.1, Center for International Comparisons at the University of Pennsylvania (CICUP), October 2002.

Per capita national income (PPP, in logarithms) Male participation rate Fitted regression line

7 8 9 10 11 0 10 20 30 40 50 60 70 80 90 100 BGD PAK PHL JAM EGY MEX RUS ZAF POL ARG KOR DNK USA CAN HKG AUS GBR GER ITA FRA BGD PAK PHLJAM EGY MEX RUS ZAF POL ARG KOR DNK USA CAN HKG AUS GBR GER ITA FRA

Per capita national income (PPP, in logarithms) Female participation rate Fitted regression line

7 8 9 10 11 0 10 20 30 40 50 60 70 80 90 100 ben40208_ch02_031-073.indd 37 ben40208_ch02_031-073.indd 37 9/8/11 10:18 PM9/8/11 10:18 PM

38 PART 1: Labour Supply

convergence occurs in the economic roles of men and women, at least as far as their participa- tion rate is concerned.

Hours

The hours-of-work aspect of labour supply has a variety of dimensions including hours per day, days per week, and weeks per year. Changes in any or all of these dimensions can alter the hours-of-work aspect of the labour supply decision. Phenomena such as the eight-hour day, the shorter workweek, and increased vacation time are institutional embodiments of a reduction in hours of work. Similarly, moonlighting, overtime, flexible working time, and compressed workweeks are institutional arrangements that alter the typical pattern of hours of work.

The policy importance of the hours-of-work decision is illustrated in a variety of ways. Changes in hours of work can affect not only the quantity but also the quality of our overall labour supply (and hence national output), as well as absenteeism, turnover, employment opportunities, and the demand for related activities—notably, those involving leisure time and flexible working hours. Changes in hours of work, in turn, can be affected by changes in the age and sex structure of the labour force, the prominence of two-earner families, govern- ment policies and laws, and, of course, the wage rate.

In the short run, hours of work appear to be relatively fixed with little scope for indi- vidual variation. The eight-hour day, five-day workweek, and fixed vacation period are fairly common for many wage and salary earners. However, the increased importance of flexible working hours is altering these arrangements. In addition, occupational choice provides an element of flexibility as people choose jobs partly on the basis of the hours of work required. Individuals may also be able to combine jobs, perhaps two part-time jobs or a full- and a part-time job, in order to work the desired number of hours for a given wage rate.

Figure 2.4 illustrates the apparent degree of flexibility that Canadian workers have in the number of hours they work per week. The work patterns of women are quite varied, while men are more likely to work the typical 40-hour week. However, even for working men, less than half work 40 hours per week. The next largest proportion (20 percent) works between 30 and 39 hours per week, while about 15 percent work more than 50 hours. Women, on the other hand, are slightly more likely to work part-time (less than 30 hours) than exactly 40 hours. Women are significantly less likely than men to work more than 40 hours per week. Might these patterns be explained by labour supply theory? Perhaps relatively higher wages for men lead to higher average hours worked, especially for the highest paid? Alternatively, women may “prefer” to work part-time so that they can better meet family commitments. Or maybe women are underpaid in the labour market, and so work fewer hours? As with participation, a good theory of labour supply should be able to accommodate these features, and the potential explanations, of labour market attachment.

As previously mentioned, the length of the workweek is also more flexible than one might think. At the beginning of the 20th century, the average factory worker worked almost 60 hours per week, the equivalent of six 10-hour days. It wasn’t until the 1960s that the average hours in manufacturing conformed to the more familiar 40-hour workweek (eight hours per day, for five days). More recently, average hours have dropped below 40 hours per week. And average hours worked are even lower, closer to the mid-30s, once we account for increased vacations and holidays.

Since real wages have risen over the century, this long-run decline in hours worked appears inconsistent with an upward-sloping labour supply function. Instead, it suggests an indepen- dent effect of increased wages: as societies become wealthier, they need not toil as hard and can afford to take more time off. It is hoped that the theoretical model will clarify the ways in which wages can affect labour supply.

ben40208_ch02_031-073.indd 38

39

CHAPTER 2: Labour Supply: Individual Attachment to the Labour Market