• Home
• About MED
• Publications
• Occasional Papers
• 08/06
• 3. Education

• Publications

• Occasional Papers

• 09/01
• 08/09
• 08/08
• 08/07
• 08/06
• 08/05
• 08/04
• 08/03
• 08/02
• 08/01
• Archive

A large body of the agglomeration literature suggests that the concentration of skilled workers is central to the productivity advantages achieved in cities. In this section we review research that has examined the effect/impact of human capital on employment growth, firm birth rates, land prices, the location decisions of firms and the location decisions of young workers among others.

Our empirical research then investigates the correlation between educational attainment and incomes in employment areas, as well as the relationship between educational attainment and sectors in employment areas in Auckland.

3.1 Literature Review

The literature consistently states that increasing the educational attainment and skills of workers, is essential to improve the economic productivity of cities. We now examine this literature in closer detail.

3.1.1 Theoretical

Moretti notes that productivity spillovers from "human capital arise if the presence of educated workers makes other workers more productive. In the presence of such spillovers, an increase in aggregate human capital may have an effect on aggregate productivity that is quite different from the effect of an increase in individual education on individual earnings".²¹ Lucas' research provides support for this, holding that human capital increases the productivity of the individual, and, at an aggregate level, contributes to the productivity of all factors of production.

The human capital literature infers that highly educated workers are the primary vehicles of knowledge spillovers.²² As Rosenthal & Strange note: "this seems to be what Marshall had in mind in his discussion of Sheffield cutlery workers taking advantage of the secrets of their trade that are available as local public goods".²³

Rosenthal & Strange, in their research on human capital spillovers, note that there is no conclusive evidence as to the mechanisms that give effect to human capital spillovers, however. They note: "learning is only one of the mechanisms by which agglomeration can impact productivity and wages".²⁴ Moretti notes moreover: "Despite the important policy implications and a large theoretical literature that assumes the existence of human capital externalities, the empirical literature on the magnitude of these externalities is still young. Given the limited number of empirical studies on this subject, it is still too early to draw definitive conclusions on the size of the externalities".²⁵

Le, Gibson & Oxley's research provides an outline of the three major approaches taken in the literature to measure human capital: cost-based, income and education. Their conclusion is that while each approach has its merits, each also has its shortcomings (missing key elements of human capital and poor data), leading to a lack of empirical consensus.²⁶

3.1.2 Empirical

The following table provides a brief summary of some key empirical studies on human capital, before a more detailed discussion is provided below.

Cooke's research is further supported by Glaeser, who remarks that the presence of skilled workers strongly influences firm location decisions and that "policy should worry about attracting people at least as much as attracting firms".³⁰ Similarly, in the economic geography literature, Simmie argues that skilled workers are a primary driver of the location decisions made by innovation led firms.³¹

Combes, Duranton & Gobillon find that up to half of the spatial wage disparities, commonly associated with agglomeration benefits, can be traced back to differences in the skill composition of the workforce. In this regard, workers with better labour market characteristics tend to sort themselves into large cities.³² Further research needs to be developed, to generate a more precise understanding of the sorting of skilled workers into cities, however. The scholarship of Nocke, and Glaeser & Maré (who point to the importance of vocational learning in cities), offer potential insights, whilst the appeal of urban amenities to skilled workers also appears to be relevant.³³

Peri's research modelled the location decisions of young and old workers as a function of human capital externalities.³⁴ Although young educated workers were found to receive a lower wage premium in urban areas, relative to experienced workers, they were over-represented in urban areas due to learning opportunities. In this respect, learning adds to human capital, which is seen to be more important for young workers than for older workers. This effect seems to be apparent here, with research indicating that young workers are attracted to Auckland.³⁵

Toulemonde's research found that firms want to settle in close proximity to high-skilled workers because they earn more money and thus stimulate demand.³⁶ Moreover, Toulemonde identified a circular causality given that workers are eager to invest in the acquisition of skills when more firms ask for skilled workers.

Rosenthal & Strange's empirical research in metropolitan areas in the U.S links different levels of worker education, and shifts in workers' educational attainment, to productivity in cities. The authors note: "proximity to college educated workers is shown to enhance productivity and wages, while proximity to less than college educated workers has the opposite effect".³⁷

3.1.3 Emerging Areas

Combes & Duranton suggest that the diffusion of knowledge within an economy is critically dependent on the "churning" of the labour force, or in other words, the movement of skilled workers from firm to firm.³⁸ The authors note: "Many case studies strongly support the idea that workers, when they change employers, come with knowledge about their former employer and that this knowledge can be profitably used by their new employer".³⁹ This builds on the earlier work of Saxenian, who examined the relationship between the churning of high skilled labour and knowledge diffusion in Silicon Valley.⁴⁰

Similar to the research of Saxenian and Combes & Duranton, an emerging branch of the agglomeration literature suggests that the economic productivity advantages of cities can be partly attributed to informal networks, where skilled workers generate and benefit from knowledge spillovers. In this regard, Storper & Venables consider face-to-face contact to be a missing aspect amongst the mechanisms known to give rise to agglomerations.⁴¹

Saxenian's analysis of Silicon Valley as a key innovation-led economy, also illustrated the significance of informal networks between skilled workers that are enabled by spatial proximity. She notes: "informal conversations were pervasive and served as an important source of up-to-date information about competitors, customers, markets and technologies… such exchanges are more able to take place in a close knit environment, where the relatively close proximity of companies makes associations easier".⁴²

Furthermore, Moretti recalls the research of Glaeser,⁴³ Peri,⁴⁴ Jovanovic & Rob⁴⁵ and Black & Henderson,⁴⁶ who, in exploring human capital spillovers in urban areas, found that individuals improve and add to their human capital through meetings with skilled neighbours where ideas are exchanged.⁴⁷

Whilst, in this phase of research, we provide no empirical examination of informal networks in the Auckland economy, an awareness of this emerging literature is helpful in providing insights into the way human capital and knowledge spillover processes may be leading to agglomeration benefits in Auckland's CBD.

3.2 Auckland Findings

Having reviewed the literature on the effects of knowledge and its transmission in supporting agglomeration, we now turn our attention to the observed position in Auckland. While there is no empirical information on the interchange of information, there is information on the educational levels of the workforce and its wage productivity on a sectoral and geographical basis. Using this information we can identify whether differences in education levels can explain, in some part, the higher productivity of workers in the CBD.

The level of educational attainment for those working in the CBD, in Auckland City and the Auckland Region is set out in Table 3.2, following the definitions used by Statistics New Zealand.

Source 2001 Census, Statistics New Zealand

Notes (1) Advanced Educational Attainment includes those with Advanced Vocational Qualifications and Bachelors and Higher Degrees

In general, the CBD demonstrates higher levels of educational attainment in its workforce, with higher proportions of workers with both bachelor and higher degrees, than either Auckland Region or Auckland City. Within the CBD, 43 per cent of employees have degrees or an advanced occupational qualification, compared to 37 per cent in Auckland City and 32 per cent in the Region. The proportion of workers with no qualification in the CBD, is also less than half of that for the Region, and significantly lower than that for Auckland City.

The differences in education levels between the CBD and the Auckland Region are also repeated at a sectoral level as can be seen in Table 3.3.

Source 2001 Census, Statistics New Zealand

Notes (1) Degree plus advanced vocational qualification

In all the sectors identified, the proportion of workers with higher educational attainment levels is higher for the CBD than for the region, even for activities like retail or construction where the educational levels are typically relatively low. The better qualified workforce may contribute to the area's productivity performance therefore.

3.2.1 Earnings & Education

The relationship between average earnings for workers with different educational qualifications is set out in Table 3.4.

Source 2001 Census, Statistics New Zealand

Higher educational levels are typically linked to higher average earnings. Within the CBD, those with some form of advanced educational qualification earn on average 22 per cent more than the average for the CBD, and those with higher degrees earn 40 per cent more than the CBD average. For Auckland, the earnings of those with a Bachelors degree are 38 per cent above the regional average, and those with a Higher Degree are almost 60 per cent above the regional average.

The higher proportions of workers with advanced educational qualifications in the CBD will be a factor contributing to the higher wage rates in the area, although, as Table 3.4 has demonstrated, earnings are higher in the CBD within particular educational categories, giving a possible indication of agglomeration benefits.

In Table 3.5 below, we set out the average earnings for workers in each of the sectors for the Region and for the Auckland CBD, and the proportions of those who have Advanced Educational Attainment. To assess the impact of the differences between those with different educational levels in the CBD and the Region as a whole the following steps were undertaken:

The average wage rate for those with and without advanced educational attainment were determined for those working within each sector in each CAU.

Revised wage rates within each sector/CAU combination were estimated assuming that the split by education type within each sector was the same as that for the Region as a whole.

Revised average wage rates were determined for the CBD as a whole by sector and in total and compared to the regional averages.

The results are set out in Table 3.5.

Source 2001 Census, Statistics New Zealand

Notes (1) AEA = Advanced Educational Attainment
** These are very small sectors

On the basis of the CBD split, the average wage rate at about $47,600 is about 29 per cent above the regional average of $37,000, with 43 per cent of the workforce in the CBD with advanced educational attainment. However, if educational attainment was in line with the regional averages for each sector, giving a total of only 32 per cent of the CBD workforce having advanced educational qualifications, wage rates at $45,800 would be about 24 per cent above the Regional average. Differences in educational attainment therefore, only account for about 17 per cent of the differences in wages rates and productivity between the CBD and the Region as a whole.

There is also an issue as to whether the higher proportion of the workforce with advanced educational qualifications within the CBD is itself an inherent component of agglomeration, rather than a separate factor. The literature outlined above, notes that skilled workers tend to congregate in cities. While controlling for educational attainment allows pure agglomeration benefits to be examined, it still leaves one to question whether the relationship between the preferences of skilled workers who predominantly choose to work in cities, and the agglomeration benefits that arise in cities, might be important.