Higher education, employment and economic growth: Exploring the interactions

Figures & data

Figure 1. Long-run unemployment rate by education: 1995–2012.

Source: PALMS, DataFirst; authors’ own calculations. Note: The 95% confidence interval is constructed by calculating the mean unemployment rate and standard deviations, and then applying the empirical rule which states that nearly all values lie within three standard deviations of the mean in a normal distribution.

Table 1. Growth and change of shares in employment, by sector: 1995 and 2012.

	None		Grades 1 to 12		Certificate		Degree		Total
	Growth	Share	Growth	Share	Growth	Share	Growth	Share	Growth	Share
Manager	−3.1	−0.1	2.5	0.2	12.1	1.1	17.4	1.3	5.6	2.5
Professionals	–	–	73.4	1.4	32.5	1.3	1.1	−0.3	7.5	2.3
Technical and associates	1.4	−0.0	2.3	0.2	–	−1.4	32.7	2.0	2.5	0.6
Clerks	−1.0	−0.0	0.1	−2.4	9.8	1.2	12.1	0.3	1.4	−1.0
Service and sales	−2.2	−0.2	4.4	2.8	8.7	0.6	13.4	0.2	4.5	3.3
Agriculture and fishing	−4.0	−0.1	−2.4	−0.5	−4.8	−0.1	−4.5	−0.1	−3.0	−0.8
Craft and trade	−2.5	−0.5	2.2	0.1	3.4	0.1	−2.7	−0.1	1.9	−0.2
Operator and assembler	−4.1	−0.9	-	−2.6	3.6	0.1	14.5	0.0	−0.3	−3.3
Elementary	−3.3	−2.9	3.4	2.8	10.3	0.2	2.1	0.0	2.2	0.4
Domestic	−3.6	−1.3	−0.3	−2.6	0.6	−0.0	-	0.0	−0.8	−3.9
Total	−3.3	−6.0	2.0	−0.6	4.5	3.0	7.5	3.3	2.1

Source: PALMS and authors’ own calculation.

Notes: The growth column presents the 17-year averaged annual employment growth, and the share column presents the shift between shares of occupation within each educational category.

Table 2. Simple output-skills elasticity: 1995–2012.

	None	Primary	Secondary	Matric	Certificate	Degree	Total
Employment	−0.1	−0.1	0.4	0.5	0.2	0.1	0.4
Labour force	−0.1	−0.2	0.3	0.3	0.2	0.1	0.4
Working age	−0.4	0.1	0.8	0.3	0.2	0.2	2.0
Population	0.2	0.1	0.3	0.2	0.1	0.2	3.1

Source: PALMS; Statistics South Africa; SARB database and author’s own calculation.

Notes: Annual data for 1995 and 2012 are used in this analysis.

Table 3. Simple two-factor Cobb–Douglas production function model (OLS results), 1995–2012.

Variable	OLS (employment)	OLS (labour force)	OLS (working age)	OLS (population)
Capital	0.543***	0.642***	0.368*	1.254***
Labour	0.639***	0.468***	0.839***	−0.425***

Source: PALMS, DataFirst; SARB database; authors’ own calculations.

Notes: ***p < 0.01, *p < 0.1.

Table 4. OLS with investment: 1995–2012.

Variable	OLS (employment)	OLS (labour force)	OLS (working age)	OLS (population)
Investment	0.219***	0.230***	0.265***	0.353***
Capital	0.555***	0.564***	0.136**	0.076
Labour	0.332***	0.293***	0.855***	0.800***

Source: PALMS, DataFirst; SARB database; authors’ own calculations.

Notes: ***p < 0.01, **p < 0.05.

Table 5. OLS with educational cohort: 1995–2012.

Variable	OLS (employment)	OLS (labour force)	OLS (working age)	OLS (population)
Investment	0.127**	0.241***	0.228***	0.245***
Capital	0.648***	0.628***	0.244***	0.13
Labour
None	0.057	0.061	0.012	0.058
Primary	−0.034	−0.054	0.125*	0.213***
Secondary	0.1	0.169*	0.688***	0.692***
Matric	0.275***	0.135***	−0.008	−0.042
Certificate	−0.135**	−0.191***	−0.069	−0.031
Degree	0.119***	0.099	0.084*	0.086**

Source: PALMS, DataFirst; SARB database; authors’ own calculations.

Notes: ***p < 0.01, **p < 0.05, *p < 0.1.

Table 6. First-stage Olley and Pakes methodology with educational cohorts.

Variable	(Employment)	(Labour force)	(Working age)	(Population)
Investment	−6.23	−2.606	−5.638	0.545
Labour
None	−0.024	−0.007	−0.09	0.029
Primary	−0.023	−0.118	−0.057	0.164
Secondary	0.145	0.280***	0.466**	0.669***
Matric	0.159	0.053	−0.023	−0.037
Certificate	−0.05	−0.075	−0.036	−0.025
Degree	0.104**	0.092*	0.102**	0.095*

Source: PALMS, DataFirst; SARB database; authors’ own calculations.

Notes: ***p < 0.01, **p < 0.05, *p < 0.1.

Table A1. Labour force growth per annum and change in share by highest educational attainment, 1994 and Q1 2012.

	OHS 1994	QLFS Q1 2012	Growth per annum (%)	Change in share (%)
None	152 983	80 299	−2.8	−6.0
Grades 0 to 7	543 064	522 424	−0.2	−16.0
Grades 8 to 11	786 797	2 087 207	9.7	6.1
Grade 12	410 549	1 508 444	15.7	12.4
Certificate	56 075	246 280	20.0	2.6
Degree	9 977	60 060	29.5	0.8
Other	12 137	29 243	8.3	0.0
Total	1 971 581	4 533 957

Source: PALMS, DataFirst; authors’ own calculations.

Notes: All changes are statistically significant at the 95% confidence interval. OHS = October Household Survey.

Table A2. Simple two-factor Cobb–Douglas production function model (OLS results) with time trend, 1995–2012.

Variable	OLS (employment)	OLS (labour force)	OLS (working age)	OLS (population)
Capital	0.703***	0.764***	0.391***	0.328***
Labour	0.374***	0.270***	0.793***	0.842***
Time	0.003***	0.003***	0.003***	0.004***

Source: PALMS, DataFirst; SARB database; authors’ own calculations.

Notes: ***p < 0.01.

Table A3. OLS with investment with time trend: 1995–2012.

Variable	OLS (employment)	OLS (labour force)	OLS (working age)	OLS (population)
Investment	0.156***	0.236***	0.236***	0.231***
Capital	0.603***	0.559***	0.164***	0.108
Labour	0.334***	0.294***	0.848***	0.889***
Time	0.001	0.000	0.000	0.002***

Source: PALMS, DataFirst; SARB database; authors’ own calculations.

Notes: ***p < 0.01.