Service labour market: The engine of growth and inequality

Alexandra Lopez-Cermeño 12 July 2015

a

A

A business absolutely devoted to service will have only one worry about profits.  They will be embarrassingly large – Henry Ford.

In the early decades of the 20th century, the frontrunners of American capitalism already envisioned the potential of the service economy. Yet, economic historians tend to explain development gaps in terms of industrialisation. Evidence from US Housing and Population Censuses challenges this classical view by revealing that the richest counties have become specialised in services rather than in manufacturing. Considering the service economy in the debate of spatial localisation suggests that market size is crucial not only for services, but for any non-agricultural activity (Desmet and Fafchamps 2005). In this column, I evaluate how the service economy triggered the evident contrast between the urban service economy and the rural US in the last century.

Specialisation, growth and inequality

The economic leadership held by the US during the 20th century increased average living standards at the cost of rising inequality across counties. US Census estimations show that average yearly wages range from $26,559 in Hancock (Georgia) to $109,405 in New York in 2010. The most prosperous hotspots are located in highly urbanised areas like New York, New Jersey, Rhode Island, Florida and California, where incomes have progressively grown with population.

Figure 1. Mean income level distribution by county

Source: Own calculations from American Community Survey 2010 (five-year estimations).

The relocation of production and people away from traditional manufacturing regions to these counties has been paired with a structural change in favour of services. According to the ILO and the Historical Statistics of the United States (2006), the US (along with other developed economies) has progressively specialised in services, increasing its share of service labour from 30 to 80% since 1890, but this trend has not been generalised across counties. The spatial match of economic activity and service employment in these hotspots suggests that the tertiary sector has not only become the motor of the US economy, but also the cause of the decay of formerly successful manufacturing regions.

Inequality within the service economy

The tertiary sector includes diverse professions such as trucking, nursing or marketing. These can be classified as:

  • Local services: – distributional and personal non-tradable services provided to individual consumers at the same time and place of production; and
  • Tradable market business services – knowledge intensive activities whose demand comes from businesses (Ciarli et al 2012).

Today, around 70% of the US labour force is devoted to local service production. Essential as they are, these services are nothing but a consequence of prosperity, while the engine of growth lies within the tradable service sector. In a recent study, Moretti (2010) estimates that an extra skilled service job generates five new positions in the local market. Part of this multiplier effect can be explained through their skill premium, which intensifies the demand of local services required by these high-salaried workers. Market size benefits local entrepreneurs in many other ways: cost and time reduction of spreading know-how, a bigger and cheaper labour market, lower prices from providers, and a larger local demand originally shaped by the multiplier effect. These forces explain the drain of workers from small and middle-size cities to highly urbanised areas.

Figure 2. Long-term locational Gini coefficients by industries across counties

Source: own calculations from US Census decennial records.

Figure 2 shows that the tertiary sector has been geographically spread during the 20th century, while manufacturing and agriculture have been more prone to agglomeration as argued by mainstream literature (Kim 1995, Krugman 1991). The study of smaller geographical units and finer industrial scales reveals this pattern is only true until the 1980s, when skilled service employees start to agglomerate more than manufacturing. The effect may seem modest because knowledge-service clusters are spread across the big cities; these are distributed across the nation and the multiplier effect mitigates agglomeration of knowledge intensive services with aggregate services, but the contrast between metropolis and rural areas is high and snowballing. As Moretti puts it, “[m]ore than traditional industries, the knowledge economy has an inherent tendency toward geographical agglomeration” (2010).

The dual economy, revisited

“We are used to thinking of the United States in dichotomous terms: red versus blue, black versus white, haves versus have-nots”, Moretti (2010) states. In this sense, the sectorial distribution of employment across the US geography illustrates the contrast and evolution between the few megalopolises that drive the total economy, and that of the rural US. Hoover’s indexes measure how disproportional the allocation of employees is in reference to the nation's distribution.

Figure 3. County Hoover’s index of primary sector employees, 1930 and 1980

Source: Own calculations from US Census records.

Figure 3 shows that most counties have a ratio of employment to agriculture higher than that of the nation. In other words, this sizeable red area represents a relatively big share of employees devoted to agriculture in low populated counties, where the land-labour relation is very high because land is abundant. Moreover, wherever the relative share of agriculture was above the national average in 1930, this specialisation was exacerbated by 1980.

Figure 4. County Hoover’s index of knowledge-intensive service employees, 1930 and 1980

Source: Own calculations from US Census records.

On the other hand, the share of high-skilled jobs seems an improbable event across the US geography according to Figure 4, where most of the counties perform below the national average. Only a few county-clusters close to big cities such as New York, Chicago, Los Angeles and San Francisco exceed the national proportion and produce enough services to cover the national needs, and even trade. Counties with a high share of skilled service employment attracted many more skilled jobs by 1980 as well.

Conveniently, the localisation of agricultural employees seems to overlap with the negative imbalance of high-skilled service jobs. Most red-shaded regions in Figure 3 are blue or white in Figure 4 because the localisation of skilled service workers is related to the presence of big local markets, which can only happen when the land-labour ratio is very low (in crowded areas). The local nature of the service economy as opposed to agriculture explains why its agglomeration has been missed by analysis of bigger geographical scales.

In other words, big cities are specialising in knowledge-intensive service sectors – becoming larger and attracting skilled and unskilled workers from smaller cities. As a result, workers from small cities are abandoning their hometowns in fear of becoming part of the rural US. County inequality is increasing, although big metropolitan areas in the upper-tail of the distribution are becoming more equal among themselves (Desmet and Fafchamps 2005).

The results of this thorough investigation show market size as a cause of localisation of any non-agricultural activity. The market effect is greater for services than for manufacturing, and it doubles for knowledge services. Meanwhile, natural endowments are most significant for agricultural production and seem to prevent the creation of big markets. This way, urban areas devoted to knowledge-based services enter a virtuous circle of growth, to the detriment of small cities and natural resource-endowed regions, which lack the effect of agglomeration economies. “In this context, initial advantages matter, and the future depends heavily on the past,” Moretti explains, “the success of a city fosters more success (…). Communities that fail to attract skilled workers lose further ground.” In other words, preconditions determine the localisation of industries, although American history has shown that economic policy can change the destiny of economies.

Policy responses to local economic decline

Industrial cities in the 1950s, like Detroit, Cleveland, and Akron, experienced an urban decay that led to the long-term slowdown of their economy. In contrast, cities like New York and San Francisco, whose markets were flooded with skilled labour, have remained prosperous and stayed at the head of the rankings of urbanisation. Big capitals offer spillovers and greater market potential, increasing the prospects of growth.

Past examples can bring hope of economic recovery to depressed areas. The state of Idaho – also known as ‘the Potato State’ – was one of the most important providers of agricultural crops of the country in the early decades of the 20th century; however average income was low and urban areas small. In the 1950s, profiting from the vast amount of unused land, the government built the National Reactor Testing Station in the nearby desert. This research centre became an international reference calling researchers to move to Idaho Falls. Average salaries rose with local demand and population. Without the government call for skilled workers to the area, Idaho Falls would have never become this prosperous.

Concluding remarks

In conclusion, knowledge services hold a self-sustaining value for the economy. Local policymakers can foster growth by attracting high-skilled workers to a region; the multiplier effect will eventually increase the local market. Nevertheless, urban prosperity creates inequality not only between urban clusters and less populated areas but also within cities. National policies could reduce the differential by reinforcing the labour market for the low skilled.

References

Broadberry, S (1998), “How did the United States and Germany overtake Britain? A sectoral analysis of comparative productivity levels, 1870-1990”, The Journal of Economic History, 58(2): 375–407.

Carter, S and R Sutch (2006), Historical statistics of the United States (Millennial Edition), Vol 2, Work and Welfare, Cambridge University Press.

Crafts, N and A Klein (2012), “Making sense of the manufacturing belt : Determinants of US industrial location, 1880-1920”, Journal of Economic Geography, 12(4): 775–807.

Ciarli, T, V Meliciani and M Savona (2012), “Knowledge dynamics, structural change and the geography of business services”, Journal of Economic Surveys, 26(3): 445–467.

Desmet, K and M Fafchamps (2005), “Changes in the spatial concentration of employment across US counties: A sectorial analysis 1972–2000”, Journal of Economic Geography, 5(3): 261-284.

Kim, S (1995), “Expansion of markets and the geographic distribution of economic activities: The trends in US regional manufacturing structure, 1860–1987”, The Quarterly Journal of Economics, 110(4): 881–908.

Krugman, P R (1991), Geography and trade, MIT press.

Moretti, E (2010), “Local multipliers”, American Economic Review: Papers & Proceedings, 100: 373-377

____________ (2013), The new geography of jobs, Houghton Mifflin Hartcourt Publishing Company, New York.

International Labour Organisation (ILO)

US Census Bureau (2010) American Community Survey, 5-Year Estimates.

_________________ Census of Population and Housing Decennial Records.

a

A

Topics:  Development Economic history

Tags:  services, manufacturing sector, US, development gap, development, history, industrialisation, knowledge services, localisation, Agriculture, regional economy

PhD candidate in Economic History, Universidad Carlos III de Madrid

Events