Wednesday, February 27, 2008

Income mobility in Armenia, 1989-2006

The contrast of the society's well being today to that of the country's pre-independence is a recurring theme. Were people better off during the pre-independence years? Did the transition years make their lot worse? Have successive governments addressed the transition pains adequately? Equally relevant, at least during this political season, is how the evolving well being of the electorate translates into votes for incumbents and competing candidates.

To address this question, I took a look at the EBRD 2006 Life in Transition Survey to study perceptions of well being now and that enjoyed in 1989. The survey asks participants to rank themselves on where they fall in the income distribution in 1989 and 2006, from the poorest 10 percent of the population to the richest 10 percent. I converted these deciles into quintiles (20 percent) and reported the resulting mobility table below (observations with missing values are dropped -- about 4 percent).

Subjective ranking: income mobility between 1989 and 2006
..1989........... 2006 Quintiles .......
Quintiles ....1.. ..2.. ..3.. ..4.. ..5. All. Households
poorest 20% 26,7% 53,0% 16,9% .1,3% 2,1% 100% 267362
next ...20% .8,8% 27,7% 58,1% .5,4% 0,0% 100% 320140
next ...20% 10,5% 34,0% 44,0% 10,3% 1,2% 100% 506791
next ...20% 11,5% 42,3% 41,0% .5,2% 0,0% 100% 720929
richest 20% 18,5% 30,0% 36,6% 10,6% 4,3% 100% 229474
.... All... 13,6% 38,0% 40,8% .6,6% 1,1% 100% 2044696
Computed from the 2006 LiTS
household survey (see data page of to download)

A striking finding is that the majority of the poorest 20 percent in 1989, the first quintile, consider themselves to be much better off in 2006. Of these 267362 households, 53 percent moved to the next quintile; 2.1 percent moved all the way up to the top quintile, the richest 20 percent. At the other extreme, of those who considered themselves the richest 20 percent in 1989, 229474 households, 95.7 percent of them became poorer by 2006; 13.6 percent moved to the bottom quintile, the poorest 20 percent.

In short, the mobility picture that emerges from above is that those who considered themselves to be the richest in 1989 are now poor, while the lot of the poor has improved considerably. I have studied income mobility using panel data for countries other than Armenia (panel data don't exist for Armenia), but I have never witnessed such (1) "rank" reversal between the poor and the rich in the income distribution and (2) mobility (only 21 percent, the sum of the diagonal, did not change rank). Of course we don't observe income, and survey participants may not correctly rank themselves (otherwise we would have about 400000 households or 20 percent in each row, for a total of 2044696 households in the last column). But nevertheless, perceptions of well being are important and should not be overlooked.

Saturday, February 23, 2008

Armenia Exit Polls 2008

The press reports that the British pollster Populus exit polls showed that 57.1 percent of the Armenian voters chose Serzh Sargsian, followed by Levon Ter-Petrosian (17.4 percent), Arthur Baghdasarian (14.6 percent), and Vahan Hovhannissian (5.98 percent); the remaining candidates received 1.6 percent. The interviews were conducted in 126 polling stations (clusters) in Armenia. These are fairly close, and within the poll's margin of error, to the official results where Sargsyan was declared the winner with 52.86% (863,000 votes).

It has been some time since I taught statistics and designed samples (not for polling), but the process of polling and extrapolating to the voting population is conceptually straight forward. First you design/select a sample with some criteria in mind to minimize sampling errors, and next apply appropriate weights to scale up the choice of those interviewed after they have voted to the size and demographics of precincts.

There are a number of problems with exit polls. For instance, the active partisan supporters of one candidate may be more likely than the supporters of other candidates to participate in exit poll interviews (selection bias). In addition, the young are more likely to participate in interviews. Furthermore, the age of the interviewers may matter as younger interviewers do not get high response rates from older voters. Of course, some of those interviewed may lie about the choice they have made (response bias).

Once the interview process is over, the next step is to weigh the sample to the size of the precincts and adjust for the non-responses say based on age and sex. One may rely here on the demographics of voters from previous elections in each precinct. While the size and demographics of the population are pretty well known (with few year lags), that of the voters is more problematic.

Exit polls are powerful indicators of corrupted elections. But without adjustments for sample selection and demographics, the results can be biased and at best meaningless. Consider the hypothetical case where most of those interviewed are old in precincts where the majority are young!

With the above in mind I was surprised to read the exit poll results from what ArmInfo reports as that conducted by the Alliance civil initiative, which I believe is organized by a large number of NGOs. Here is the full article:
Exit poll by Alliance civil initiative gives victory to Levon Ter-Petrossyan
2008-02-19 20:54:00
ArmInfo. Alliance civil initiative has conducted an exit poll at 100 polling stations. They have questioned 4,406 people. 3,550 of them (66%) agreed to answer.

The vice chairman of Alliance Gevork Melikyan says that ...
474 people (15.5%) voted for Artur Bagdassaryan,
40 (1.3%) for Artashes Gegamyan,
38 (1.2%) for Tigran Karapetyan,
8 (0.2%) for Aram Haroutyunyan,
195 (6.4%) for Vahan Hovhannissyan,
90 (2.9%) for Vazgen Manukyan,
9 (0.2%) for Arman Melikyan,
1084 (35.4%) for Serzh Sargsyan,
1152 (37.7%) for Levon Ter-Petrossyan.

Melikyan says that it was an independent survey. They planned to question 0.2% of all voters - 2,300 people.
This is strange. The sample design and selection seem odd. And what are we to make of the sample sizes of 2300, 4400, 3550, and the 3090 (sum of the people voting for individual candidates). Why are the sample estimates (1152, 1084, ...) reported without any adjustment for the size of the voting population (precincts), and seemingly not any of the demographics? What do the above figures tell us about the likely winner of the elections? Clearly, and unless ArmInfo got it wrong somehow, at best this is sloppy work. None of the foreign news services, except for Radio Free Europe (RFE/RL), reported its findings. Instead, they reported those of the reputable Populus.

Monday, February 18, 2008

Remittances and education

Sometime in October of 2007 I read the job market paper of a PhD student from UC Santa Cruz, Migrants' Remittances and Investments in Children's Human Capital: The Role of Asymmetric Preferences in Mexico, which addresses how remittances affects spending on education by the recipients. Using household survey data for Mexico, matched to migrants in the US, the paper finds that the impact of remittances depends largely on the gender of the household head, and they are most likely invested in education particularly if they are sent by migrant fathers to mothers.

Last month the IMF released a new working paper entitled "Microeconomic Implications of Remittances in an Overlapping Generations Model with Altruism and Self-Interest" which addresses the impact of remittances on education expenditures, as well as on labor supply, saving, and borrowing from banks in Armenia. It is authored by two US trained PhDs originally from Armenia but residing in the US. Using the 2004 Integrated Living Standards Measurement Survey conducted by the National Statistical Service of Armenia (Armstat), and in contrast to the study using Mexican data, the authors report that in Armenia "remittance-receiving households ... spend less on the education of their children."

Unlike the Mexican migration data (download from here), the Armenian household survey data is not publicly available. And unfortunately, the IMF authors do not provide any summary statistics on the separate attributes of households with and without remittances. Other than the reported econometric estimates, there is no way to verify or question the validity of the findings. Indeed, it is very difficult to evaluate the intertemporal effects of remittances using cross sectional data, add to this the endogeneity of much that is observed in the data (e.g. housing quality which authors use as proxy for wealth). Also on practical grounds, and in the Armenian context, what does it mean to say that remittance recipients spend less on educating their children? Primary and secondary education is compulsory, private elementary and secondary schools don't exist for the most part, and ... So how are they reducing spending on education (never mind the cross sectional nature of the data and the econometric problems this creates).

As an alternative, I resort to the EBRD Transition in Life Survey (LiTS) 2006 survey data which contains some 1000 Armenian households (check data page to download). Unlike the 2004 household survey, the LiTS 2006 survey identifies the recipients of remittances but not the size of the transfers. Below are sample statistics on each group, restricted to households with children age 6 through 21 -- variable names age3-age12 in the data.

Households with remittances seem to spend more on education; an average of USD 200 per year more than the households not receiving remittances (389 vs 184). They are more likely to be headed by a woman (45 vs 39 percent), more likely to have a college degree, less likely to receive social benefits from the government, and more likely to receive gifts and transfers from others in the country.

I also estimated an OLS regression of the determinants of education expenditures. Consistent with the basic statistics above, households receiving remittances (dummy), spend USD 243 more on education. However, when remittances are the most important source of income, households spend only an additional USD 88 (i.e. 243 - 155). The qualitative results don't change when the observations are weighted, estimated for those reporting education expenditures only, and/or for the entire sample of 1000 observations without age restrictions. Of course this analysis is very simplistic in that it does not control for the endogeneity of remittances or migration, and is plagued with LiTS data limitations. Nevertheless, it would be useful for the IMF study, when revised in the future, to report basic statistics on the households in the 2004 sample and inform the rest of us on the various attributes of households in Armenia.

If you are interested in studying the attributes and behavior of Armenian households, whether you are an economist, a social scientist, and/or a political junkie, you will find the LiT survey very interesting -- in one example it allows you to compare 2006 to 1989. It also allows you to compare households in Armenia to others in all transition economies. Best of all, it is free. But please do keep the pressure on governmental agencies and NGOs to make their survey data available. CRRC is a good role model here.

Friday, February 15, 2008

Is this the best way to provide technical assistance?

I have been concerned for some time about the lasting effects of technical assistance. USAID, EBRD, among many others provide funding for training in many aspects of governance as well as the transition of the country to a market economy. These agencies, and the donor countries in general, have been quite generous. However, a common feature of most of these programs is that universities are by-passed in this process, both in terms of training faculty as well as funding that is devoted to set up new bureaucracies where the training is to take place.

Take the recent announcement for actuarial training to commence in April. There are no actuaries to speak of in the country. And so it is difficult to imagine how the insurance industry will take off without the skilled labor. As such the country is truly lucky to have USAID fund a new program to train actuaries in the country. Granted that this will be the first time such training is made available, but this begs the question of why this is not taking place in the mathematics, statistics, or even the business schools at the various universities in Yerevan who are desperate for training and funding? What happens when the funding is over? Who trains the next generation? Who undertakes research on future developments in the various underlying markets? Who will write the future textbooks?

Faculty salaries are low by any measure, and textbooks in various subjects of actuarial and management sciences, finance, and economics, among many others, do not exist for the most part. Donors have been very generous, but one would hope that they would take steps to more aggressively engage academia, that is if they wish to leave a lasting legacy.

Saturday, February 02, 2008

The Armenian Railway Deal -- more is needed

Recently a Russian firm acquired the rights to manage the Armenian railroad system. The Railway Gazette provides a good summary of the background leading to this transaction, with more detail on the Armenia Rail Concession web site. The system has experienced serious reduction in service and is in desperate need of funding.

Cargo and passenger travel, in millions

In general, and for a variety of reasons, the government and the public seem uncomfortable with privatizing publicly owned assets. At the same time, the public enterprises seem unable to raise capital. The public-private partnership model, where ownership remains with the government but management is delegated to a second party, is perhaps a good compromise. In addition to the railways, other examples include Zvartnots airport (Argentina) and the water system (France), among others.

It is wonderful to see the government go through the railways deal. But this raises the question of why it is taking so long for the government to embark on such transactions. The slow privatization process since 1991 has run many productive assets to the ground, and landed others in the hand of oligarchs. Add to this the fact that foreign investments in the form of FDI are small relative to the country's needs as well as to that taking place in the neighboring countries. And so let's hope more such transactions take place.