The Index of Prestige: Measuring socioeconomic status in Albania
Measuring wealth and socioeconomic status has always been a difficult task in Albania, when considering its economic characteristics. A big percentage of the economy is based on informality, and wage workers make up a minority of the labor force, in contrast to most developed countries. A large percentage of the workforce is also supposed to be employed in agriculture (around 35–40%), although the reliability of such figures is low. Another big source of income are remittances, money brought to the household from family members working abroad.
However, new data from wages, occupations, and cars leaked in 2021, combined with the 2008 civil registry, can help provide a relatively plausible and accurate way to measure socioeconomic status. The calculation employed here uses household wage income, occupational and locational prestige, and cars to calculate a final index ranging from 0 to 100, where a score like 50 corresponds to a status better than 50% of the population and worse than the other 50%.
Methodology
To calculate the final index, we need four variables that I managed to create and compute based on the available data:
- Occupational prestige
- Household wage income
- Locational prestige
- Vehicle prestige
Using the leaked wage data from 2021, we have each person’s occupation in Albania based on the ISCO-08 classification system. To rank occupations according to their prestige, I used three measures: the mean surname age of people in each occupation, their average family size, and their average wages. The latter doesn’t always work, as some occupations, like dentists, do not have fixed wages. So, I use family size and mean surname age as proxies for socioeconomic status, as there is a strong correlation between lower fertility and higher education in Albania. People with surnames having a high mean age signify lower fertility and likely higher life expectancy. The family size average provides further information about fertility rates and family types for each individual, and finally, average wages play a crucial part as well.
Based on the 2008 civil registry, I can calculate household size and measure wage per capita within each household, although this measure has limitations due to not having births after 2008 and affecting some age groups more than others. Then, I use wage data and occupational distribution to measure the prestige of each address in Albania. Areas with higher wages and a higher percentage of people employed in managerial and professional occupations receive higher scores. Finally, I again use these measures to create a vehicle prestige list.
With all four variables in place, I calculate them for all individuals believed to be living in Albania. For someone to be considered living in Albania, they must live in a household where at least one person has a wage, a vehicle, or where a child was born after 1997. This results in a population of around 2.55 million inhabitants. For those without occupations, I calculate their occupational prestige as the average occupational prestige of the household members with an occupation. The same is done for cars. Finally, I use a scaled PCA (principal component analysis) to create a final index that is then converted to a percentile score ranging from 0 to 100.
Results
While the methodology is complex and undoubtedly has biases, the results have been generally satisfactory and pass the smell test. We can use a variety of variables to check the median index and understand the main results. Examining addresses, we see that the top results are located in upscale neighborhoods in Tirana:
Number 1 on the list is what is unofficially known as Rruga e Salës, where ex-PM Sali Berisha lives, and is an area known for its nice cafes and restaurants. The rest of the top 10 mostly consists of neighborhoods in the Blloku area, where leaders of the communist regime lived before 1991 and access was restricted to outsiders.
As for occupations, this is the top 10 list:
Judges, politicians, managers, and some highly specialized professionals make up the top 10 list. We can also check the least prestigious occupations:
Most of the bottom 10 occupations are related to mining, shoemaking industry, and agriculture. These are usually occupations with lower wages or high physical risks. Nobody with good resources would accept to work in these occupations.
If we look at the companies and workplaces with the highest achieved index scores, we see that they are mostly related to foreign companies or institutions:
The index also allows us to examine other factors tied to history, such as the region of origin, religion, and ethnicity of individuals. This map shows the median prestige score by settlement of origin:
There is a strip of settlements in the southernmost parts of Albania that make up most of the top list. These mostly coincide with areas that also had higher literacy in 1945, although there is good performance in areas such as Tepelene and Skrapar as well. The latter regions had a mediocre level of development in 1945 and no major urban center but participated at a high rate in the communist movement, reaping its results. In the rest of Albania, we see green parts mostly in the urban areas (Tirana, Durres, Shkoder, Elbasan), as well as in Tropoje, which might be explained by the fact that Sali Berisha, leader of Albania in the periods 1992–97 and 2005–13, came from this region. Thus, while there is continuity, some regions have also profited from the historical political environment.
Dissecting this further with religion and ethnicity, the top 10 groups are as follows:
Most of the top 10 is led by the Orthodox of southeastern Albania (Gjirokaster, Kolonje, Permet, Korce, Pogradec), which is related to an earlier level of development in this community due to higher literacy and a lower percentage working in agricultural occupations. Gjirokaster is the best performing region overall, with its main towns of Gjirokaster and Libohove well known for their production of elites during the Ottoman Empire and Gjirokaster being the birthplace of Enver Hoxha, the dictator of Albania during 1944–1985.
The bottom 10, on the other hand, is mostly composed of groups from Northern and rural Central Albania:
A curious fact is the underperformance of Catholics in Shkoder. This population group has historically been the source of many intellectuals, writers, and other members of the Albanian elite. However, the rural population of this group has always suffered from isolation and a low level of development. This, combined with the repression of the Catholic elite and its institutions by the communist regime, might explain the causes of this underperformance.
We can also check the regional origin distribution of each percentile in our index:
In the top 1%, southerners and people from urban Tirana (the capital) make up most of the population. It should be noted that a relatively high percentage of people from urban Tirana are themselves of southern origin, so the overall percentage of southerners by descent in the top 1% could be close to 65%.
In terms of religious background, we see this distribution:
I have separated the overall Muslim groups from the secularized Muslims, meaning those individuals where none of the family members born after 1945 have had Muslim first names. We can observe how their percentage increases in the top 1% compared to the rest of the Muslims by background. This highlights the importance of secularization as an elite marker. A spatial analysis of Tirana neighborhoods also shows a very high correlation between a higher percentage of Muslim first names and lower socioeconomic status.
Finally, the score also reveals class differentiations in first names, where we see a marked preference among the elites (especially after 1990) for shorter names.
Conclusions
Exploring the results of this index and relating it to other variables and historical indicators could fill up a whole book, but I believe I have shown here the main results. To further validate my model, I also used the microdata from the 2011 census and created a similar socioeconomic index using years of education and household asset possessions (car, internet connection, microwave, etc.). The results and trends were relatively similar when compared between the same variables.
Nonetheless, more data would improve our index. Unfortunately, we don’t have any data on wealth or income apart from wages. In a way, this wealth would show up in car possession and the address of individuals, but it would be even better to have direct data on this. What would be even harder is to measure income from informal activities and remittances from abroad. Thus, the index can be considered quite accurate for urban areas (particularly Tirana), but less so in the rest of the country.
In the interest of transparency and self-analysis, I applied the index to measure my own score, resulting in a 98.2, which is not too bad. Should you have an interest in learning the score of any individual, do not hesitate to ask — I am prepared to provide the respective Index of Prestige score upon request.
