BSE Voice – Page 27 – Economic analysis and reflections from students and alumni of the Barcelona School of Economics

Variance: regression, clustering, residual and variance – Liyun Chen ’11

Liyun Chen ’11 (Economics) is Senior Analyst for Data Science at eBay. She recently moved from the company’s offices in Shanghai, China to its headquarters in San Jose, California. The following post originally appeared on her economics blog in English and in Chinese. Follow her on Twitter @cloudlychen

Variance is an interesting word. When we use it in statistics, it is defined as the “deviation from the center”, which corresponds to the formula , or in the matrix form (1 is a column vector with N*1 ones). From its definition it is the second (order) central moment, i.e. sum of the squared distance to the central. It measures how much the distribution deviates from its center — the larger the sparser; the smaller the denser. This is how it works in the 1-dimension world. Many of you should be familiar with these.

Variance has a close relative called standard deviation, which is essentially the square root of variance, denoted by // . There is also something called the six-sigma theory– which comes from the 6-sigma coverage of a normal distribution.

Okay, enough on the single dimension case. Let’s look at two dimensions then. Usually we can visualize the two dimension world with a scatter plot. Here is a famous one — old faithful.

Old faithful is a “cone geyser located in Wyoming, in Yellowstone National Park in the United States (wiki)…It is one of the most predictable geographical features on Earth, erupting almost every 91 minutes.” We can see there are about two hundreds points in this plot. It is a very interesting graph that can tell you much about Variance.

Here is the intuition. Try to use natural language (rather than statistical or mathematical tones) to describe this chart, for example when you take your 6 year old kid to the Yellowstone and he is waiting for next eruption. What would you tell him if you have this data set? Perhaps “I bet the longer you wait, the longer next eruption lasts. Let’s count the time!”. Then the kid has a glance on your chart and say “No. It tells us that if we wait for more than one hour (70 minutes) then we will see a longer eruption in the next (4-5 minutes)”. Which way is more accurate?

Okay… stop playing with kids. We now consider the scientific way. Frankly, which model will give us a smaller variance after processing?

Well, always Regression first. Such a strong positive relationship, right? ( no causality…. just correlation)

Now we obtain a significantly positive line though R-square from the linear model is only 81% (could it be better fitted?). Let’s look at the residuals.

It looks like that the residuals are sparsely distributed…(the ideal residual is white noise which carries no information). In this residual chart we can roughly identify two clusters — so why don’t we try clustering?

Before running any program, let’s have a quick review the foundations of the K-means algorithm. In a 2-D world, we define the center as // , then the 2-D variance is the sum of squares of each pint going to the center.

The blue point is the center. No need to worry about the outlier’s impact on the mean too much…it looks good for now. Wait… doesn’t it feel like the starry sky at night? Just a quick trick and I promise I will go back to the key point.

For a linear regression model, we look at the sum of squared residuals – the smaller the better fit is. For clustering methods, we can still look at such measurement: sum of squared distance to the center within each cluster. K-means is calculated by numerical iterations and its goal is to minimize such second central moment (refer to its loss function). We can try to cluster these stars to two galaxies here.

After clustering, we can calculate the residuals similarly – distance to the central (represents each cluster’s position). Then the residual point.

Red ones are from K-means which the blue ones come from the previous regression. Looks similar right?… so back to the conversation with the kid — both of you are right with about 80% accuracy.

Shall we do the regression again for each cluster?

Not many improvements. After clustering + regression the R-square increases to 84% (+3 points). This is because within each cluster it is hard to find any linear pattern of the residuals, and the regression line’s slope drops from 10 to 6 and 4 respectively, while each sub-regression only delivers an R-square less than 10%… so not much information after clustering. Anyway, it is better than a simple regression for sure. (the reason why we use k-means rather than some simple rules like x>3.5 is that k-means gives the optimized clustering results based on its loss function).

Here is another question: why do not we cluster to 3 or 5? It’s more about overfitting… only 200 points here. If the sample size is big then we can try more clusters.

Fair enough. Of course statisticians won’t be satisfied with these findings. The residual chart indicates an important information that the distribution of the residuals is not a standard normal distribution (not white noise). They call it heteroscedasticity. There are many forms of heteroscedasticity. The simplest one is residual increases when x increases. Other cases are in the following figure.

The existence of heteroscedasticity makes our model (which is based on the training data set) less efficient. I’d like to say that statistical modelling is the process that we fight with residuals’ distribution — if we can diagnose any pattern then there is a way to improve the model. The econometricians prefer to name the residuals “rubbish bin” — however it is also a gold mine in some sense. Data is a limited resource… wasting is luxurious.

Some additional notes…

Residuals and the model: as long as the model is predictive, then residuals exist, regardless of the model’s type, either a tree or linear or whatever. Residual is just the true Y minus the prediction of Y (based on training data set).

Residuals and loss function： for ordinary least squares, if you solve it in the numerical way then it iterates by the SSR (sum of squared residuals) loss function (equals to the variance of residuals). In fact many machine learning algorithms relay on a similar loss function setting — either first order or higher order moments of residuals. From this perspective statistical modelling is always fighting with residuals. This differs from what the econometricians do so there was a huge debate on the trade off between consistency and efficiency. Fundamentally different believes of modelling.

Residuals, Frequentists and Bayesians: In the above paragraphs I mainly followed the Frequentist’s language. There was nothing on posterior… From my understanding many items there would be mathematically equivalent to the Bayesian’s frameworks so it should not matter. I will mention some Bayesian ideas in the following bullets so go as you wish.

Residuals, heteroscedasticity and robust standard error: We love and hate heteroscedasticity at the same time. It tells us that our model is not perfect while there is a chance to make some improvements. Last century people tried to offset heteroscedasticity’s impact by introducing the robust standard error concept — Heteroscedasticity-consistent standard errors, e.g. Eicker–Huber–White. Eicker–Huber–White changes the common sandwich matrix (bread and meat) we use for the significant test (you may play with it using the sandwich() package in R). Although Eicker–Huber–White contributes to the variance estimation by re-weighing with estimated residuals, this approach does not try to identify any patterns from the residuals. Thus there are methods like Generalized least square (GLS) and Feasible generalized least square (FGLS) that try to use a linear pattern to reduce the variance. Another interesting idea is clustered robust standard error which allows heterogeneity among clusters but constant variance within each cluster. This approach only works when the number of groups approaches infinite asymptotically. (otherwise you will be getting stupid numbers like me!)

Residuals and reduction of dimensions: generally speaking the more relevant co-variates introduced to the model the less the noise is; while there is also a trade-off towards overfitting. That is why we need to reduce the dimensions (e.g. via regularization). Moreover, it is not necessary that we want to make a prediction every time; sometimes we may want to filter out the significant features — a sort of maximizing the information we could get from a model (e.g. AIC or BIC or attenuation speed which increasing the punishment in regularization). In addition regularization is not necessarily linked to train-validation… not the same goal.

Residuals and experimentation data analysis: heteroscedasticity will not influence the consistency of Average Treatment Effect estimation in an experimentation analysis. The consistency originates from randomization. However people are still eager to learn more beyond a simple test-control comparison, especially when the treated individuals are very heterogenous; they look for heterogenous treatment effect. Quantile regression may help in some case if there is a strong covariate observed…but what could we do when there are thoudsands of dimensions? Reduce the dimension first?

Well, the first reaction to “heterogeneous” should be variance…right? otherwise how could we quantify heterogeneity? There is also a bundle of papers that try to see whether we would be able to find more information for treatment effects rather than simple ATE. This one for instance:

Ding, P., Feller, A., and Miratrix, L. W. (2015+). Randomization Inference for Treatment Effect Variation. http://t.cn/RzTsAnl

View full code in the original post on Ms. Chen’s blog

Monitoring the Spanish Economy: “Spain’s response to EC and OECD economic policy recommendations”

alumni Víctor Burguete ’11 (International Trade, Finance and Development) is an Economic Researcher and Public Policy Analyst at IESE’s Public-Private Sector Research Center (IESE-PPSRC) in Barcelona. In this post, he shares the process of preparing a policy brief on Spanish policy reforms and provides an overview of the brief’s findings.

Preparing a Policy Brief like this took me over a month. It is necessary to consider than working in a research institution implies getting involved in many projects and there is usually less time than what I would like to devote to one specific project. In my opinion, it is very important to work open-minded and to continuously consider the possible connections among different projects. In the case of this Policy Brief, most of the data (international economic policy recommendations) were collected during the past few months. In late September I proposed this topic and the IESE-PPSRC research center decided to inaugurate these series of papers. After reviewing the literature (Table 1), I analyzed the data and I started creating some graphs and building the story I wanted to tell. Of course, the final text was reviewed several times until it was finally published.

“Spain’s response to EC and OECD economic policy recommendations” analyses the overall reformist progress of the Spanish Government in an international perspective. According to the international assessment, Spain ranks as one of the top reformers in the Euro Area and the EU as a whole. A second insight one gets from our Policy Brief is that Spain’s delivery, in relative terms to other countries, accelerated between 2011 and 2013.

Of course, this is the general trend and the Policy Brief offers details on the progress in the 18 policy sub-areas we cover at the SpanishReforms project, including how the reform priorities prescribed to Spain by these institutions have changed over time. Substantial progress is recognized in addressing the financial system reform, mainly in the area of recapitalization and restructuring but also by adopting other financial measures. However, both the OECD and the EC point to active labour market policies and professional services as the main structural reforms lagging behind.

More information in www.spanishreforms.com, a new an academic, non‐governmental website that aims at being a useful reference for those interested in independent, rigorous and up‐to‐date information about the Spanish economy and its economic policy reforms.

The Death of Fixed Income

Alex Hansson ’13 (International Trade, Finance and Development) is an Analyst at Tribus Capital Partners in Zurich, Switzerland. Previously he was External Asset Management Analyst at Credit Suisse.

A few weeks ago I found myself sitting in the James Joyce Wine Cellar in Zurich listening to a talk by the Chief Investment Officer of one of the world’s largest asset managers’ fixed income division. Surrounded by ageing wine bottles and Swiss bankers in grey suits, the CIO put down his glass of Chateau Neuf du Pape and proclaimed, “You know, I’ve been investing in fixed income all my career, but I feel obliged to tell you that fixed income as an asset class is dead.” That was the last sentence of the speech.

Now I don’t frequent too many of these events, but ending a presentation on that kind of a bombshell was not something that I’d ever experienced before. The Swiss bankers around me shifted nervously in their seats thinking something else was coming. But the CIO did not follow up the statement with any caveats. He was undoubtedly out to provoke and had we pushed him a little he would have most likely followed up his statement with the usual spiel that one should be careful with statements like ‘this time it is different.’ He would have probably added that fixed income in terms of developed markets sovereign and corporate bonds would at some point in the future again be an interesting investment.

Nonetheless, finding out fixed income was dead was pretty exciting. Moreover, the statement was in line with a trend I had been witnessing for the past couple months whereby capital was flowing into a new asset class – alternative fixed income.

But, before moving on to this, it is worth spending a few minutes on the death of fixed income, whether temporary or not. Classic portfolio management theory will tell you to put 60% of your assets in equities (e.g. at its most basic through a S&P 500 Index ETF) and 40% in bonds (e.g. at its most basic through a Bond Index) in order to achieve diversification and lower the portfolio’s volatility. Without getting too deep into the technicalities, the thinking goes that equities outperform bonds in the long run – hence the overweight. However, bonds are included because they lower the volatility and increase the diversification of the overall portfolio. They lower the volatility because they are less risky than equities. Moreover, bonds have historically moved inversely to equities and this diversification effect has served the purpose of smoothing out the return profile of a balanced portfolio in times of market turbulence. Seeing as, nearly all investors have regular liquidity needs for, for instance, a regular mortgage payment, a smoother albeit lower return profile in the long run is more attractive than high volatility with a higher return in the very long run.

So what has changed? Norman Villamin (CIO Coutts) was recently quoted as saying that bonds used to offer an asymmetric payoff relative to equities whereas now you are left with a “symmetric asset class.” Put another way, bonds still serve the purpose of protecting capital when equities are down but no longer give you any meaningful upside. In other words by having 40% bonds you lower volatility but you no longer achieve diversification. Explaining exactly why, at least from a developed markets perspective, is not straightforward. Some analysts have highlighted that rates have been kept low to encourage lending given the slower than expected recovery in the US and Europe. Other analysts have pointed to lower inflation and central banks not feeling the pressure to increase rates. In any case, with very low returns you also have very low volatility. With low volatility even if this effect moves inversely to equities it will not be ‘strong enough’ to provide the desired counterbalancing effect. Think of two grown men on a seesaw where one gets replaced by a child, no matter how much the child jumps up and down on the plank, they will not significantly propel the man into the air. Whatever the reasons for this phenomenon, most seem to agree that a low rate environment will be the norm for some time to come.

Many investors realize that if rates remain low it will become more difficult to find bonds yielding attractive returns without taking on excessive risk. In terms of fixed income this generally means looking towards emerging market debt or distressed corporate bonds. To be clear some investors have successfully gone down this route, such as, Michael Hasenstab (CIO Franklin Templeton) who has became famous for his bullish speeches on Ukrainian debt. Nonetheless this dynamic has led to many investors leaving fixed income in favor of equities. That trend is clearly reflected by the fact that the largest fund by assets under management has over the last year shifted from the PIMCO Total Return fund (a fixed income product) to the Vanguard Total Stock Market Index (an equity product). The leaving of Mr. Gross from PIMCO no doubt also contributed to this shift.

With the recent prolonged bull market most investors have shut their eyes to the increased volatility and enjoyed the attractive returns they have seen from their portfolios. However, some sophisticated investors have elected not to move their fixed income exposure to equities. Some have chosen to put assets into well-known alternatives like private equity or real estate. But these asset classes do not offer the same advantages that fixed income did.

As outlined above fixed income was liked because, first, it was relatively low risk. Second, it paid a steady regular coupon, which was liked not just by private investors but especially pension and insurance funds, which have regular liquidity needs. Third, it paid a handsome return of anywhere from 5.00% to 8.00% (historically seen). To put that into perspective, a ten year German Government Bond pays you less than 1.00% currently. A private equity or real estate investment conversely can mean having to lock up capital for ten years, with no regular coupon payment, and with no guarantee for handsome reward.

Given this, many of these sophisticated investors have turned to a different asset class that has emerged largely as a product of new lending opportunities. It is often referred to as Alternative Fixed Income. These opportunities have to a large extent emerged as a product of new banking regulations. Simply put, banks have been forced to move away from certain types of lending and new non-bank credit providers have stepped in to fill this void offering new investment opportunities. Indeed, according to Alliance Bernstein, since 1980 the nonfinancial corporate and mortgage credit outstanding has grown by ca. USD 18.9tn whereof USD 15.0tn can be attributed to non-banks.

This trend of lending by non-banks has been accelerated largely as a product of post-crisis regulation. In Europe, the most important regulation framework for this trend has been Basel III. This framework stipulated that banks now have much higher capital requirements as well as having to take on higher operational costs for certain types of lending. This has been most pronounced for investments with longer investments horizons, which therefore cannot match bank-liability structures subject to daily liquidity requirements.

There are literally hundreds of structures that have emerged to take advantage of this systemic shift. The biggest opportunities for alternative credit providers have been in corporate loans, commercial real estate loans, residential real estate loans, and infrastructure loans. Many new organizations have been set up and are raising capital from investors in order to lend to corporations and real estate developers in the same way banks used to. The lenders pay a coupon on the loan, the new firms take a cut, and the investor gets a regular coupon payment at a tolerable risk and a much higher return than they would have in traditional fixed income.

This systemic shift has created opportunities along the entire supply chain of lending. It stretches from the most logical to the most niche areas you can think of, and no doubt this is just the beginning. Below I will give two examples of the most common constellations.

Most commonly you see new organizations made up of former bankers. They leave or are made redundant by their employer. They take with them an intimate knowledge of that bank’s loan book as well as the skillset to analyze other banks’ loan books. They then raise capital in a fund structure (funded by our sophisticated investors above) and begin buying loans from banks that can no longer be kept on their balance sheets. This can take many different forms. The easiest is a pure purchase at a discount where the new organization takes over the loan. Typically this happens via a Collateralized Loan Obligation (CLO) which is a securitized asset backed by a pool of debt. Another variety is a situation where banks have debt on their balance sheet which is highly attractive from a return perspective but which they cannot keep on their balance sheet. They therefore use something called Regulatory Capital Relief Trades (CRTs) in order to temporarily transfer risky assets off their balance sheets to one of these new organizations which in turn charge interest on the assets while on their balance sheets. Finally you have situations where the bank will give an organization access to their pipeline of loan opportunities. This setup is rarer and requires a very close relationship between the bank and the new organization. Essentially, these new organization can sift through bank’s pipeline of lending opportunities and chose to take on any loans they find attractive. In return the bank receives a share of the profits.

The other more common constellation of new organizations is direct loans that banks used to be able to do. Indeed, many of these organizations are whole teams that used to do the same thing at their former employers. A great example is Renshaw Bay. Renshaw Bay was setup by the former co-CEO of J.P. Morgan’s Investment Bank – Bill Winters. The firm runs a real estate strategy that is “focused on direct, whole loan origination of commercial real estate loans… and seeks to take advantage of the lack of financing available to real estate borrowers”. They also run a structural finance strategy, which looks to “capitalize on opportunities driven by regulatory change and the retreat of capital.” There are now new organizations (or old organizations which have seen a significant uptick in demand) that are doing the same thing in corporate loans, real estate loans, infrastructure loans, or even leverage loans to private equity firms.

It is still relatively early days and no one yet completely understands the full implications of this systemic shift. Perhaps the setup is better, seeing as you have more specialized niche players, often with much of their internal capital at risk, and unhindered by bureaucracy associated with banks running these new organizations. Or perhaps this setup is riskier, since at the same time you also have small, inexperienced, and at times highly leveraged organizations with less oversight then banks used to have. Added to this increasing numbers of pension funds and insurance companies are investing in these structures. Undoubtedly there will be some financial cowboys who have cut corners in order to have first mover advantage. At the same time you will undoubtedly have a lot of smart people who will make themselves and their investors a lot of money. Clearly, it is going to be very interesting to monitor these developments to see whether this shift turns out to be a good one or a bad one.

The Mission: Human Capital and the Persistence of Fortune – Job Market Paper

Job market paper Felipe Valencia ’15 (GPEFM – UPF and Barcelona GSE)

The following job market paper summary was contributed by Felipe Valencia (GPEFM – UPF and Barcelona GSE).

**Update: This paper has now been published in the Quarterly Journal of Economics and featured in the The Washington Post!**

The importance of history in economic development is well-established (Nunn 2009; Spolaore and Wacziarg 2013), but less is known about the specific channels of transmission which drive this persistence in outcomes. Dell (2010) stresses the negative effect of the mita in Latin America, and Nunn and Wantchekon (2011) document the adverse impact of African slavery through decreased trust. But did other colonial arrangements lead to positive outcomes in the long run?

I address this question in my Job Market Paper by analyzing the long-term economic consequences of European missionary activity in South America. I focus on missions founded by the Jesuit Order in the Guarani lands during the seventeenth and eighteenth centuries, in modern-day Argentina, Brazil and Paraguay. This case is unique in that Jesuits were expelled from the Americas in 1767 –following European “Great Power” politics— precluding any continuation effect. While religious conversion was the official aim of the missions, they also increased human capital formation by schooling children and training adults in various crafts. My research question is whether such a one-off historical human capital intervention can have long-lasting effects.

The author at the site of one of the Jesuit missions on Guarani lands in South America.

Setup

To disentangle the national institutional effects from the human capital shock the missions supplied, I use within country variation in missionary activity in three different countries:

fig1_valencia — Note: The map shows the exact location of the Guarani Jesuit Missions (black crosses) with district level boundaries for Argentina, Brazil and Paraguay.

The area under consideration was populated by a single semi-nomadic indigenous tribe, so I can abstract from the direct effects of different pre-colonial tribes (Maloney and Valencia 2012; Michalopoulos and Papaioannou, 2013). The Guarani area also has similar geographic and weather characteristics, though I control for these variables in the estimation.

Key Findings

Using municipal level data for five states (Corrientes and Misiones in Argentina, Rio Grande do Sul in Brazil, and Itapúa and Misiones in Paraguay), I find substantial positive effects of Jesuit missions on human capital and income, 250 years after the missionaries were expelled. In municipalities where Jesuits carried out their apostolic efforts, median years of schooling and literacy levels remain higher by 10-15%. These differences in educational attainment have also translated into higher modern per capita incomes of nearly 10%. I then analyze potential cultural mechanisms that can drive the results. To do so I conduct a household survey and lab-in-the-field experiments in Southern Paraguay. I find that respondents in missionary areas have higher non-cognitive abilities and exhibit more pro-social behavior.

Endogeneity

Even though I use country and state-fixed effects as well as weather and geographic controls, Jesuit missionaries might have chosen favorable locations beyond such observable factors. Hence the positive effects might be due to this initial choice and not to the missionary treatment per se.

To address the potential endogeneity of missionary placement, I conduct two empirical tests. The first one is a placebo that looks at missions that were initially founded by the Jesuits but were abandoned early on (before 1659). I can thereby compare places that were initially picked by missionaries with those that actually received the missionary treatment. I find no effect for such “placebo” missions, which suggests that what mattered in the long run is what the missionaries did and not where they first settled.

Second, I conduct a comparison with the neighboring Guarani Franciscan Missions. The comparison is relevant as both orders wanted to convert souls to Christianity, but Jesuits emphasized education and technical training in their conversion. Contrary to the Jesuit case, I find no positive long-term impact on either education or income for Franciscan Guarani Missions. This suggests that the income differences I estimate are likely to be driven by the human capital gains the Jesuits provided.

In addition, I employ an IV strategy, where I use as instruments the distance from early exploration routes and distance to Asuncion. Distance from the exploration routes of Mendoza (1535-1537) and Cabeza de Vaca (1541-1542) serves as a proxy for the isolation of the Jesuit missions (in the spirit of Duranton et al. 2014). Asuncion, in turn, served as a base for missionary exploration during the foundational period, but became less relevant for Rio Grande do Sul after the Treaty of Madrid (1750) transferred this territory to Portuguese hands. For this reason and to avoid the direct capital –and Spanish Empire—effects, I use this variable only for the Brazilian subsample of my data (as in Becker and Woessmann 2009; Dittmar 2011). The first-stage results are strongly significant throughout (with F-statistics well above 10), and the second-stage coefficients for literacy and income retain their sign and significance –appearing slightly larger—in the IV specifications.

Extensions and Mechanisms

To complete the empirical analysis, I examine cultural outcomes and specific mechanisms that can sustain the transmission of human capital from the missionary period to the present. I find that respondents in missionary areas possess superior non-cognitive abilities, as proxied by higher “Locus of Control” scores (Heckman et al., 2006). Using standard experiments from the behavioral literature, I find that respondents in missionary areas exhibit greater altruism, more positive reciprocity, less risk seeking and more honest behavior. I use priming techniques to further investigate whether these effects are the result of greater religiosity –which appears not to be the case.

In terms of mechanisms, my results indicate that municipalities closer to historic missions have changed the sectoral composition of employment, moving away from agriculture and towards manufacturing and services (consistent with Botticini and Eckstein, 2012). In particular, I document that these places still produce more handicrafts such as embroidery, a skill introduced by the Jesuits. People closer to former Jesuit missions also seem to participate more in the labor force and work more hours, consistent with Weber (1978). I also find that indigenous knowledge —of traditional medicine and myths—was transmitted more from generation to generation in the Jesuit areas. Unsurprisingly, given their acquired skills, I find that indigenous inhabitants from missionary areas were differentially assimilated into colonial and modern societies. Additional robustness tests suggest that the results are not driven by migration, urbanization or tourism.

Follow Felipe on Twitter

Young and under pressure – Europe is risking a lost generation

The following post by Olga Tschekassin ’13 (Master in International Trade, Finance and Development) has been previously published by the World Economic Forum and Bruegel.

Ms. Tschekassin is Research Assistant at Bruegel in Brussels, Belgium. Follow her on Twitter @OlgaTschekassin

Since the beginning of the global financial crisis, social conditions have deteriorated in many European countries. The youth in particular have been affected by soaring unemployment rates that created an outcry for changes in labour market policies for the young in Europe. Following this development, the Council of Europe signed a resolution in 2012 acknowledging the importance of this issue and asking for implementation of youth friendly policies in the Member States. Yet, almost 5.6 million young people were unemployed in 2013 in the European Union (EU) – in nine EU countries the youth unemployment rate more than doubled since the beginning of the crisis.

Today I want to draw your attention to two more indicators reflecting the social situation of the young generation: the percentage of children living in jobless households and the percentage of young people that are neither in employment nor education nor training.

Children in jobless households

The indicator Children in jobless households measures the share of 0-17 year olds as a share of the total population in this age group, who are living in a household where no member is in employment, i.e. all members are either unemployed or inactive (Figure 1).

Source: Eurostat and Bruegel calculations. Country groups: 10 other EU15: Austria, Belgium, Denmark, Finland, France, Germany, Luxembourg, Netherlands, Sweden and United Kingdom; Baltics 3: Latvia, Lithuania, Estonia; 10 other CEE refers to the 10 member states that joined in the last decade, excluding the Baltics: Bulgaria Czech Republic, Croatia, Hungary, Poland, Romania, Slovenia, Slovakia, Cyprus and Malta; Sweden: data for 2007 and 2008 is not available, the indicator is therefore assumed to evolve in line with the other 9 EU15 countries. Such approximation has only a marginal impact on the aggregate of the other EU15 countries, because children in jobless HHs in Sweden represented only 3% of the country group in 2009. Countries in groupings are weighted by population.

In the EU28 countries this share rose only slightly over the past years to 11.2%. It is striking, however, that the ratio of children living in households where no one works more than doubled in the euro-area programme countries (Greece, Ireland, Portugal) as well as in Italy and Spain to 13% and 12%, respectively. And even more shocking – while the share stabilized in the programme countries, in Italy and Spain it is still sharply increasing. In Ireland in 2013 more than one in every six children lived in a household where no one worked. This is indeed an alarming development. Only the Baltics, which experienced a very deep recession among the first countries hit by the crisis, are reporting a sizable turning point in the statistic in 2010 and the share is presently continuing to decline. The numbers are, however, still well above pre-crisis levels.

A high share of children living in jobless households is not only problematic at the moment but can also have negative consequences for the young people’s future since it often means that a child may not only have a precarious income situation in a certain time period, but also that the household cannot make an adequate investment in quality education and training (see a paper on this issue written for the ECOFIN Council by Darvas and Wolff here). Therefore a child’s opportunities to participate in the labour market in the future are likely to be adversely affected. Moreover, as I discussed in a blog earlier this year, children under 18 years are more affected by absolute poverty than any other group in the EU and the generational divide is widening further.

Not in Education, Employment or Training (NEET)

The financial situation of young people between 18 and 24 years old who finished their education is less dependent on their parents income because they usually enter the labour market and generate their own income. Therefore we are going to have a closer look on their work situation, i.e. how many young people have difficulties participating in the labour market.

Figure 2: Not in Education, Employment or Training

Source: Eurostat and Bruegel calculations. Country groups as in previous chart

The NEET indicator measures the proportion of young people aged 18-24 years which are not in employment, education or training as a percentage of total population in the respective age group. We can see in Figure 2 that the situation among EU28 countries stabilized over the last four years. The good news is that for the first time since 2007 we see a decline in the rate in the euro-area programme countries in 2013. This decline is, however, mostly driven by Ireland with an unchanged situation in Greece and Portugal. Also, in the Baltics the ratio is on a downward trend. More worrying, however, is the situation in Italy and Spain. Among all EU28 countries, the young generation in Italy with 22.2% of all young people being without any employment, education or training, is disproportionately hit by the deterioration in the labour market. Every fifth young person between 18 and 24 is struggling to escape the exclusion trap. Europe and especially Italy is risking a lost generation more than ever.

Labour market policies for young people should therefore stand very high on the national agendas of Member States. The regulations introduced in summer 2013 into the Italian labour market reform which are setting economic incentives for employers to hire young people build an important step towards more labour market integration of the youth in Europe. Their effects are yet to be observed in the employment statistics in the coming years in Italy. More action on the national and European level is needed to improve the situation of the young.

The global network of payment flow – Barcelona GSE Data Scientists

Originally posted by Jordi Zamora ’15 on the Barcelona GSE Data Scientists blog.

At the Rényi Hour on November 20th, Samantha Cook presented her recent research on the description and categorisation of the global SWIFT (Society for Worldwide Interbank Financial Telecommunication) interbank network. Samantha is currently the Chief Scientist at Financial Network Analytics in Barcelona. Previously, she was a Quantitative Analyst at Google’s Research Group in New York and a professor at Columbia University in New York and Pompeu Fabra University in Barcelona.

φωτογραφία 2 — Samantha Cook ready to give her talk

The study focused on understanding the underlying structure of a network of messages between financial institutions in different countries. It looked at how the network was affected by various recent economic events and evaluated the robustness of the system over time.

The data set underpinning the study contains standard MT103 SWIFT messages from 1 January 2003 and 31 July 2013, a period characterised by extreme economic turmoil. Each message represents a single customer credit transfer from bank to bank. The data is aggregated at the country level.

Samantha showed us different statistical analyses of the data set. The analysis of the data in terms of a complex weighted network was particularly interesting. In the network, each node represented a country and the edges connecting two different nodes were weighted according to the amount of messages those country exchanged in a given time period. The resulting network follows approximately a Core-Peripheral structure, that is, some nodes are fully connected with each other (the so-called core) while some others are mostly connected only to a node of the core: these are the peripheral nodes. Interestingly, events such as the introduction of new regulations or the beginning of the financial crisis was clearly reflected in the links and even more striking this network structure was resilient during the period studied. This work showcases a novel approach to understanding the structure of the complex financial system and the findings may provide a way to help improve the global service.

The discussion also identified some opportunities for further research. For example, we discussed why the degree distribution does not behaves as other related financial networks, and why the number of links decreases while the number of messages has a clearly increasing trend. These questions, and others that emerge, may provide ideas for further research and modelling work in this area.

Useful links:

Heterogeneous Inputs, Human Resource Management and Productivity Spillovers: What Do Poultry Farm Workers Have to Say? – Job Market Paper

The following job market paper summary was contributed by Francesco Amodio (Economics ’10 and GPEFM). Francesco is a job market candidate at UPF. He will be available for interviews at the SAEe (Palma de Mallorca, December 11-13) and ASSA (Boston, January 3-5) meetings.

Management matters. Differences in management practices can explain a considerable amount of variation in firms’ productivity and performance, both across and within sectors and countries (Bloom and Van Reenen 2007, 2010, 2011). Several studies have shown how human resource management and incentive schemes may affect overall productivity by making the effort choices of coworkers interdependent (Bandiera, Barankay and Rasul 2005, 2007, 2009). In more complex settings, however, workforce management features may interact with production arrangements and jointly determine the overall result of the organization. Understanding the nature of this interplay is of primary importance in the adoption and implementation of productivity-enhancing management practices.

In my job market paper, coauthored with Miguel A. Martinez-Carrasco, we shed light on these issues by focusing on settings where workers produce output by combining their own effort with inputs of heterogeneous quality. This is a common feature of workplaces around the world. For instance, in Bangladeshi garment factories, the characteristics of raw textiles used as inputs affect the productivity of workers. Similarly, the purity level of chemicals affects the productivity of researchers in biological research labs.

Now, suppose we pick a worker and endow her with higher quality inputs, thus increasing her productivity. What happens to the productivity of coworkers around her? Do they exert more effort, or do they shirk? How do human resource management features shape their response?

The setting

In order to answer these questions, we collected data from an egg production plant in Peru. Production is carried out in production units located one next to the other in several sheds. In each production unit, a single worker is assigned as input a batch of laying hens. Workers’ main tasks are to feed the hens, to maintain and clean the facilities, and to collect the eggs. The characteristics of the hens and worker’s effort jointly determine productivity, as measured by the daily number of collected eggs. Figure 1 shows the picture of one shed hosting four production units. Notice how workers in neighboring production units can easily interact and observe each other.

The specific features and logistics of this setting generate the quasi-experiment we need in order to answer the questions of interest. All hens within a given batch have very similar characteristics. When reaching their productive age, they are moved to one production unit and assigned to the corresponding single worker who operates the unit. After approximately 16 months, they reach the end of their productive age and are discarded altogether. The age of hens in the batch exogenously shifts productivity. Indeed, Figure 2 shows the reversed U-shaped relationship that exists between hens’ age and productivity. Perhaps more importantly, the timing of batch replacement varies across production units, generating quasi-random variation in the age of hens assigned to workers.¹ We can thus exploit these differences to credibly identify the causal effect of an increase in coworkers’ productivity – as exogenously shifted by coworkers’ hens age – on own productivity, conditional on own hens’ age.

Main Results

We find evidence of negative productivity spillovers. The same worker, handling hens of the same age, is significantly less productive when coworkers in neighboring production units are more productive, with variation in the latter being induced by changes in the age of their own hens. This finding is pictured in Figure 3, which shows that a U-shaped relationship exists between own productivity and coworkers’ hens age. In other words, workers exert less effort and decrease their productivity when coworkers are assigned higher quality inputs.

We also find similar negative effects on output quality, as measured by the fraction of broken and dirty eggs collected over the total number of eggs. Furthermore, we find no effect of an increase in the productivity of coworkers located in non-neighboring production units or in different sheds, suggesting that workers only respond to observed changes in coworkers’ productivity.

The role of HR

Why do workers exert less effort when coworkers’ productivity increases? Our hypothesis is that the way the management processes information on workers’ productivity in evaluating them and taking employment termination decisions generates free ride issues among coworkers. When observed productivity is only a noisy signal of workers’ exerted effort, the management combines available signals and best guesses the level of effort exerted by the worker. Even when observable input characteristics can be netted out, individual signals are still imperfect, and possibly excessively costly to process. The management thus attaches a positive weight to aggregate or average productivity in evaluating a single worker. As a result, workers free ride on each other.

In order to test for this hypothesis, we collected employee turnover data from the same firm. As expected, we find that the likelihood of employment termination is lower the more productive the worker is. More importantly, being next to highly productive workers improves a given worker’s evaluation and diminishes her marginal returns from effort, yielding negative productivity spillovers.

We also find that providing incentives to workers counteracts their tendency to free ride. First, we find no effect of coworkers’ productivity when workers are exposed to piece-rate pay. Second, we collected data on the friendship and social relationship among workers, and find again no effect of coworkers’ productivity when a given worker recognizes any of her coworkers as friends. We interpret this as further evidence that the main result of a negative effect of coworkers’ productivity indeed captures free riding issues, mitigated by the presence of social relationships.

Discussion

Our focus on production inputs and their allocation to working peers represents the main innovation with respect to the previous literature on human resource management and incentives at the workplace. In our case study, the allocation of inputs of heterogeneous quality among workers triggers free riding and negative productivity spillovers among them, generated by the workers’ evaluation and termination policies implemented at the firm.

The analysis of more complex production settings reveals the existence of intriguing patterns of interplay between production arrangements and human resource management practices. Our plan for the next future is to proceed further along this line of inquiry. In a companion paper still work in progress, we investigate both theoretically and empirically how workers influence each other in their choice of inputs while updating information on the productivity of the latter from own and coworkers’ experience.

¹ Grouping all observations belonging to the same shed and week and taking residuals, we show that the age of hens assigned to coworkers is orthogonal to the age of own hens. We test this hypothesis in several different ways, addressing the issues arising when estimating within-group correlation among peers’ characteristics (Guryan, Kroft, and Notowidigdo 2009; Caeyers 2014). We cannot reject the hypothesis of zero correlation in all cases.

Can big data be official? – Barcelona GSE Data Scientists

Originally posted by Stefano Costantini ’15 on the Barcelona GSE Data Scientists blog. Stefano is on Twitter @stefanoc.

Originally posted by Stefano Costantini ’15 on the Barcelona GSE Data Scientists blog. Stefano is on Twitter @stefanoc.

At the Renyi Hour on November 13th 2014, Frederic Udina gave a talk on big data and official statistics. Apart from being a professor at UPF and BGSE, Frederic is Director of IDESCAT, the statistical institute of Catalonia.

Frederic Udina presenting to BGSE Data Science students

In his talk, Frederic compared the “traditional” official statistics – slow to produce, with well-defined privacy limits and access rights – to “big data”, which is fast to produce, volatile and with fuzzy privacy limits. Frederic highlighted the tension between these two worlds, focusing particularly on the need for official statistics to become easier to collect, organise and customise to the need of the final user. In particular, Frederic identified the opportunity for IDESCAT (and other statistical institutes) to integrate the officially collected information with alternative information sources, such as:

Administrative data
Data freely available from the society
Data from private companies

Frederic outlined IDESCAT’s plan to move away from the current data generation system (the ‘stove pipe model’) which is slow, expensive and inefficient as it does not re-use information already collected, towards a fully integrated model (‘Plataforma Cerdà’) where any new information needs to be integrated with existing data.

Frederic noted that data is becoming increasingly important in society, and this is beginning to be recognised by official statistical institution. In particular, Frederic discussed the Royal Statistical Society’s Data manifesto where the RSS notes that data is:

A key tool for better, informed policy-making
A way to strengthen democracy and trust
A driver of prosperity.

The Royal Statistical Society Data Manifesto

Frederic also stressed the importance of confidentiality and privacy issues with regards to data availability. While it is desirable for some data to be freely available to the public, confidentiality and privacy should always be protected. However, it is important to strike the right balance between access and privacy, ensuring that while personal sensitive data is protected, important information is not prevented from being used in ways that may ultimately help the wider society. Personal health records are a classic example of this.

Frederic concluded his talk by providing some example of national statistical authorities integrating official statistics with widely available information to carry out new interesting analysis. Examples include:

Production of origin/destination arrays between territorial units (usually municipalities) for working or studying reasons using trajectories of mobile phones (ISTAT, New Zealand Statistics)
Using Google Trends to estimate/predict labour market, monthly forecast, small-area estimation (ISTAT)
Measuring use of TCI in firms, by using web scraping and text mining techniques

Useful links:

Breakfast seminars: food for thought

By Marlène Rump ’15, current student in the International Trade, Finance and Development master program at Barcelona GSE. Marlène is on Twitter @marleneleila.

On Wednesday, October 22, we didn’t have classes, so we decided to explore one of the numerous events on the GSE calendar. For some brain and other food, the breakfast seminar on Labour, Public and Development Economics sounded just right.

The presentations scheduled were held by two of UPF’s PhD students who are in their last year. This means they are finalizing their “job market paper”, which refers to the paper they will use as a demonstration of their skills and interests when they apply for positions.

One important purpose of the seminar is giving the students an opportunity to practice presenting and defending their work, as well as receiving improvement suggestions from fellow PhD students and professors.

Backlash: The Unintended Effects of Language Prohibition in US Schools after World War I

Vicky Fouka started the seminar with her paper on language prohibition in the US Schools after World War I. She compared two states, similar in most social aspects, one of which banned the teaching of German from the primary schools for a few years and the other, her control state, which didn’t.

The prohibition, which was implemented by the authorities in early 1920s, originated from a German-hatred which was widespread in the United States after World War I. What was promoted as an integration measure had the exact opposing effects: Vicky finds that the Germans living in the state with language prohibition deepened their cultural segregation. In comparison with the control state, they were more likely to marry a German spouse and give their first child a very German sounding name.

Editor’s note: Vicky Fouka is a graduate of the Barcelona GSE Master in Economics. See more of her research on her website.

Cultural Capital in the Labor Market: Evidence from Two Trade Liberalization Episodes

The second presentation was also about the assimilation of immigrants, however Tetyana Surovtseva conducted her analysis with modern day data. Her assumption was that if the host country of immigrants increased trade with their country of origin, these immigrants had an advantage on the labor market in trade related sectors. Her hypothesis was that if the host country of immigrants increased trade with their country of origin, these immigrants had an advantage on the labor market in trade related sectors. Her underlying premise is that immigrants have a certain “cultural capital”, other than language, which is valuable for corporations involved in trade with their country of origin.

Tetyana examined the labor market demand for Chinese and Mexican immigrants in the US after a punctual improvement of trade agreements. Her findings suggest that labor market returns to the immigrant cultural capital increase as a result of trade with the country of origin.

Editor’s note: Tetyana is also a Barcelona GSE Economics alum. More about her work is available on her job market page.

Attend some seminars! Especially if you’re thinking of doing a PhD.

For both presentations there were numerous questions which gave additional insight especially on the methods of research. We also learned that most PhD students start their final thesis three years before the end of their program.

After this experience, I can highly recommend attending the seminars. You learn about interesting economic questions and see a specific application of your econometrics classes and this in only one hour. In addition, for those who are envisaging doing a PhD, the presentations give a genuine insight of the type of research you could be conducting.

Are capital markets the only friend of innovation?

Post by Barcelona GSE alum Marco Antonielli ’12 (Master in International Trade, Finance and Development)