Search This Blog

Friday, March 30, 2012

Industrial Structure and Culture - A Brief View

I would have expected the structure of small isolated economies to resemble the nearest largest economy, but Oceania (dominated by Australia) and the Caribbean do not. Instead, they resemble the largest economic group that shares their cultural background regardless of distance.

In the Caribbean, the largest island economies are primarily Spanish speaking and as a group their structure closely resembles Spanish speaking Latin America. This situation is not completely clear, but complicated by two major twists. The economic structure of English speaking Jamaica also matches the Latin American structure, so resemblance based on cultural background does not always hold. In Puerto Rico’s case, its full economic integration with the US renders it irrelevant to this issue.


Fully one half of Australia’s bilateral trade is with OECD countries, which seems strange given that the OECD group is the geographically most distant 
(this trade data is from Australia Department of Foreign Affairs and Trade http://www.dfat.gov.au/publications/tgs/index.html ).


Source: derived from UN online SNA database (System of National Accounts)

I am defining industrial structure as the proportion of GDP contributed by the main 3 types of industry: agriculture, manufacturing and services. In the graph, I have shown industrial structure as represented by the share of agriculture alone. This is because agriculture is much better correlated with GDP per capita than the other sectors, and so allows a simple graphical representation.
Using agriculture also seems intuitively valid because, as the oldest human industry, the less reliant on it an economy is, the further its economy has developed.

For more information on Gravity Theory as it relates to culture and history, see an excellent paper by Douglas Campbell (UC Davis), www.econ.ucdavis./campbell.pdf

Monday, March 26, 2012

How Many More Low Paying US Jobs Exist After the Recession?

Weakness in the US labour market persists into the 4th year of the recession with the most intense sectoral shift hitting manufacturers. The gradual decline in the manufacturing share of GDP continues a decades-long trend which has accelerated since 2008. Out of 12.6 million production jobs in 1999, 4.4 million jobs were lost by 2010 so there is no doubt about the severity of the structural shift.

A common feature of recessionary job losses is the claim that low paying service sector jobs replace high paying factory jobs. That is, burger flipping McJobs instead of unionized and skilled work. In addition to the recession, this sounds like what we would expect given the rise of China and increasing globalization. It seems reasonable to think that importing manufactured goods would leave fewer jobs, and that those would be of lower value; we should also be able to find out how many workers have been affected in this way.

There are many complaints about the way un-employment figures are generated. These are focused on stringent definitions of job seekers that end up creating an overly rosy view of the labour market. Since un-employment figures can be misleading, I have compiled Employment data for every job category from the US Bureau of Labor Statistics. 

The most recent data is from 2010, well into the recession, while the earliest data of the same classification system is from 1999. (Of course the downside of using employment data is that people will complain about ignoring population growth; there is no avoiding that, but it is still worth looking at this half of the story).

The idea is to address one specific question: how many high paying jobs were lost and replaced by low paying jobs? To do this, I listed the major job categories and ranked them by their 1999 pay level. Then I divided the total 1999 workforce into 2 segments of equal size, high pay and low pay. Finally, I counted the number of jobs in each of the 2 segments in 2010. This gives the number of jobs gained or lost in each pay segment. (please see the table below to help visualize the strategy)

The Results

Production job wages ranked nearly centred between the high and low halves of the workforce. The surprise here was that the production category is not in the high paying half of the workforce. Instead, it is ranked second in the lower wage segment, just below the “protective service” category that includes police officers, prison guards and firefighters. 

It seems that the high pay of production jobs is at least partly a myth, while the popular image of low paying food service jobs is confirmed by its last ranked position in the list. The 11 million food service jobs are indeed low paying service sector jobs. But as a counter-balance, the 11 million service sector jobs in engineering, computers and business are high paying.

Out of approximately 127 million jobs in 1999, the total change by 2010 was a loss of 176, 840 jobs ( 0.14 %). But instead of the expected decline in higher wage jobs, the exact opposite occurred. More than 1.6 million low wage jobs were lost, and more than 1.4 million high wage jobs were added.

Production jobs showed the largest absolute loss of any category at nearly 4.4 million. More surprisingly, the next largest job category change in was in management with a loss of just over 2 million. This includes a 54% loss of CEO positions and a 66% loss of advertising management jobs.
As a proportion of total jobs in each category, production workers and managers lost the same amount, 25% since 1999.
The biggest job gaining areas in the high paying category were business & financial operations, and health care practitioners.

Below is the categorized jobs data.



Source: table derived from http://www.bls.gov/oes/oes_dl.htm
accessed 25 March 2012

Sunday, March 18, 2012

Does the Embargo Affect Cuba's GDP?

It seems reasonable to suspect that the embargo would have a crushing impact on Cuba's output, so this effect should show up in a comparison of Cuba with its closest neighbours. Most trade theorists would agree that cutting off the largest and closest trading partner will seriously reduce a small economy's performance (usually indicated by GDP per capita). This does not seem to have happened, because Cuba is right in line with its regional peers, Jamaica and Dominican Republic.

The UN is virtually the only data source on Cuban GDP as the government does not consistently publish this figure. I graphed each country's real per capita output using UN data, and added Puerto Rico even though it is fully integrated with the US economy (not really a trading partner so much as another state).


Cuba - red; Jamaica blue; Dominican - green

 
While Puerto Rico is alone at a much higher level, the other three islands form a tight group. Cuba starts the period (1970 - 2008) midway between Dominican Republic and Jamaica but ends the period ranking first. Dominican Republic grows along with Cuba and only Jamaica fails to progress over the 38 years.


It seems reasonable to expect that Cuba’s worst circumstances for growth would be under the embargo along with no majority trading partner. And yet in the most recent period, with exactly those conditions and a tightened embargo, Cuba has seen its highest growth. 

This argues against the idea of economic harm from restraint of trade. Jamaica is free to trade with everyone in the world, has an English language advantage in US markets and is even a member of the Commonwealth since 1962, and yet has seen a steadily deteriorating growth rate for over half a century. Dominican Republic and Puerto Rico are more typical in that they had fluctuating positive growth over the decades although Puerto Rico clearly has an economic advantage in being fully integrated with the US.
 
The structural shift from agriculture to manufacturing and eventually to services is associated with higher long term GDP growth in countries across the Caribbean and across the globe. In other words, less agriculture is strongly correlated with higher national income. The issue for Cuba is that it fits so very neatly into this world-wide economic pattern of modest income during a structural shift away from agriculture. This means that the idea of unique and significant harm to Cuba from the American embargo is highly implausible. Its economic performance matches so many other American trading partners with a similar industrial structure that the embargo clearly has not singled out Cuba from its regional or global cohorts.

So the ultimate question regarding the impact of the embargo is whether Cuba's economy would have soared if it had been allowed to continue selling sugar and cigars to the US for decades. The global experience shows that trade with the US would not suffice. Likewise trade in agricultural and manufactured goods with any other country would not suffice. Instead it seems that Cuba could have fared better had it transformed its economy more quickly toward the international export of services.