Tuesday, January 5, 2016

Flags and economics

Last month we saw on the news the New Zealanders (or Kiwis) getting involved in a referendum to choose a new national flag.
New Zealand confirmed that a blue, white, red and black fern and stars design won the referendum to become the contender of the current flag. A second referendum will be held in March to decide whether to adopt the new flag or keep the old one.

No research, as far as I know, analysed before the relation between flag designs and economic or political performance (all I found is this). That is because there is no point whatsoever in doing that. Any relation should be spurious or the consequence of an omitted variable bias (i.e. the assumed specification is incorrect in that it omits an independent variable that is correlated with both the dependent variable and one or more included independent variables). Nonetheless I thought it could be a light-hearted albeit interesting analysis.

I run three regressions. In the first one I used the flag design to predict the Quality of the Democracy (as measured by the Economist Intelligence Unit) by country. The second regression dependent variable is the logarithm of the GDP per capita and the third one is the Economic Complexity as measured by The Observatory of Economic Complexity of the MIT. The reference in all three regressions is a flag with the red colour in it, horizontal stripes and no symbols (those were, independently, the most common features).

Interestingly, having some sort of cross in a country flag has the largest positive impact on that country’s democracy. Adding some kind of weapon such as machine guns (Mozambique) or swords has the largest negative impact. A substantial amount of green colour does, as well, have a negative impact.

In GDP pc terms, the most negative impacts come from having a star or other symbols and a dominant green or yellow colour. On the other hand the positive and statistically significant impact comes from having a moon on the flag.

Finally, the economic complexity, a good predictor of economic development and future performance, is positively correlated with a cross and the colour white. Conversely, it is negatively correlated with the colour green and stars.

In summary, this analysis would suggest that New Zealand would be better off if their flag include some kind of cross and a significant amount of white colour. They should avoid at all cost adding a slash, or weapons, using the green colour and above all, get rid of the stars!

Friday, December 4, 2015

Public sector private sector

It is a common believe among European citizens that public sector workers or civil servants are overpaid and enjoy too many privileges. In theory, one could argue that they are likely to be so, in most countries they are highly unionised and their industrial actions can produce havoc and huge disruptions on basic and essential services. Thus, through their lobby and influence they could force governments to accept higher than market wages. This is, however, just theoretically. The main problem with this theory is that in order to know if someone is overpaid or swamped with privileges, we need to know first the value of their services. How can we know the value of a service that is not traded and has no market price?

One could compare public wages to private sector wages on similar services and industries. But is that enough for a fair comparison? What other factors should one consider and how should one measure them? This is what Fabien Postel-Vinay from UCL tries to do on this paper.

What he finds is that using data for the period 1994-2003, “direct wage comparisons show that public-sector employees earn [on average] around 15% more than private-sector employees.” 

However, two adjustments need to be considered. “First, the public and private sectors differ in the jobs they offer and the type of workers they employ. Second, public- and private-sector careers also differ in other important dimensions, such as job stability and income progression, which are relevant to individual career choices”.

Any comparison of the public-private gap should take these points into account.

The first point made by Postel-Vinay is quite sensible and self-evident. We need to compare like for like and it is widely documented that the public sector tends to attract better educated and more experienced workers than the private sector. This is why the author takes from the private sector only those workers that have the experience and education similar to the average public worker.

The second bit, however, is where the study derails, the author says: “Finally, the lifetime job values used in the […] comparisons assume that workers never change sectors or experience unemployment.” A private sector with a long term zero unemployment is wishful thinking.

The graph below shows the Public-private gap in lifetime job values by percentile.

We can amend those values using unemployment rates for the period 1994 to 2014 from Eurostat and using the differences in % of workers who moved from public sector to unemployment compared to the % who moved from private sector to unemployment shown in Figure 1 of the same paper.

The results are the following:

Tuesday, October 20, 2015

Economic history

How the study of history affects economics research field?

The study of the past was neglected by economists in the early 80’s. Deirdre McCloskey surveyed the use of economic history in the 70’s and 80’s and she found that economists “believe history to be of small and diminishing interest”, she concluded that the average American economist answers “no” to the question “Does the past have useful economics?” McCloskey showed a sharp decline in the publication of economic history papers in the top economic journals (AER, QJE, JPE).

Today, economic history has reverted that trend and, although still small –compared to other fields, it has become a relevant field of study according to most of research economists.

Ran Abramitzky published last month a paper about the use of economic history by economists. He found that the top 5 economic journals have increased their share of papers related in one way or another to economic history.

This idea that the past influences the present in a path dependence process is now widespread. Many economists argue that a process of path dependence often takes place in the modern world: norms and expectations impede change, discrimination survive even in highly competitive markets, and change can be very slow.

Another point Abramitzky makes in his paper is the problem with endogeneity in economic history. Identification of a causal effect is a main challenge for economic history but that’s no different from applied economics. Over the last two decades researchers have started to take causal identification more cautiously, changing their language to more lightly words that imply but not state causality such as “effect”, “impact” and “influence,” and only claiming causality when a random or quasi-random variation is established.

Economic history has increased the use of data and quantitative tools to solve the endogeneity problem. The classic example is the AJC 2001 paper.

Abramitzky ends quoting Arrow “it will always be true that practical understanding of the present will require knowledge of the past.”

Saturday, September 5, 2015

Race against the machine

The global labour markets are constantly changing and the key demanded skills change with those. In the latest years we have seen a rising concern about the role of computers in the labour market. We have obviously seen these before (e.g. Luddites) but many say this time might be different. Machine learning and computer learning is replacing human cognitive tasks of rapidly increasing complexity.

These pattern has been analysed by many, among those; Frey and Osborne in 2013 , Wolff in 2005, and Benzell et al.

A new academic paper published in early August by David Deming tries to identify the recent trends on skills’ demand and builds a theoretical model that predicts the changes since the 80’s. It's a compelling story. The most interesting finding in the paper is the following:

“Nearly all job growth since 1980 has been in occupations that are relatively social skill-intensive. Jobs that require high levels of analytical and mathematical reasoning but low levels of social interaction have fared especially poorly.”

The argument goes like this: as more computers replace human analytical power the relative value of social skills become more important. Computers are still very poor at simulating human interaction. Reading the minds of others and reacting accordingly seems to be a very complex thing to do.

Sunday, August 9, 2015

Peru’s mining mita

Reading the notable book from Acemoglu and Robinson, "Why nations fail" I came across the concept of mita and the effect of it on economic growth. Mita was an extensive forced mining labor system in effect by the Spanish Empire in some regions of today´s Peru and Bolivia between 1573 and 1812. Fascinatingly, a paper wrote in 2011 by Melissa Dell (or here) found that:
Regression discontinuity results indicate that a mita effects lowers household consumption by 25% and increases the prevalence of stunted growth in children by around 6% points. Mita’s influence has persisted through its impacts on land tenure and public goods provision. Today, those regions are less integrated into road networks and their residents are substantially more likely to be subsistence farmers and with lower educational attainment.
Undoubtedly, it’s been known that historical institutions are key on current economic outcomes. This paper not only confirms this statement but finds the routes through which those institutions can have this persistent effect on growth. Aka public goods provision and land tenure. The key element of this research is that the mita required indigenous communities to send a constant share of their adult population to work in the silver and mercury mines and conscripts changed discretely at the boundary of the subjected region: on one side, all communities sent the same percentage of their population, while on the other side, all communities were exempt. This is probably an arbitrary and short informed decision of the regional government ef the Spanish Empire that allows the researcher to use as an external shock.

Friday, July 31, 2015

Machine learning visual experiment

R2D3 is an experiment in expressing statistical thinking or machine learning with interactive visual design. In machine learning, computers apply statistical learning techniques to automatically identify patterns in data. These techniques can be used to make highly accurate predictions.
In this case the authors Stephanie and Tony design  a machine learning process in order to know if a house with unknown location is from San Francisco or New York based on other parameters such us altitude, price, sq feet,…

In most cases they seem to apply a logistic regression where high or low values of one variable such as high altitude are more likely to be from one city rather than the other. By sequentially applying this process with all variables and using recorded data the authors can get a very accurate estimate of location for any given house, provided the explanatory variables are available.

Saturday, July 4, 2015

Coloured time series

Here comes an idea (I saw it in Flowingdata.com) to graph many time series together in a well-designed and simple way.

The thing is that when you try to put many graphs together, the Y-axis becomes too short and changes are really hard to appreciate.

These kind of graphs can be converted into colour graphs which are much easier to understand. The Y-axis is fixed to a given level for all graphs. If one of the time series goes above that level, a darker area start at that point from the Y=0. Negative numbers can be drawn as reddish.

It is much clearer when a time serie is up or down by looking at colours.