Usage Of R in Defining Labour Market Areas

Labour Market Area is a territory in which high rate of people both live and work. It does not need to be consistent with area restricted by administrative boarders. It seems rather obvious that administrative boarders are not always a proper criterion in making certain decisions. The government should sometimes base on functional regions instead. That is the reason for which in 2013 Eurostat decided to from the Task Force responsible of creating a common methodology of defining Labour Market Areas. A couple of member states have already had certain experience in using their own definitions of Labour Market Areas. Nevertheless, due to independence of each method the results achieved in different countries were obviously incomparable. This caused a need to create a methodology, which would be universal for entire European Union. According to proposal there are two criteria deciding whether an area is a Labour Market Area or not – size (number of employed inhabitants) and self-containment, which is a minimal value of the following: 1) the proportion of an area’s employed population that works within the area and 2) the proportion of jobs within an area that are filled by residents of that area. Since the results should be possibly stable, population censuses are desirable sources. As travel-to-work matrices contain relations between places of living and places of work at LAU-2 level, datasets may have large size (for Poland almost 350 000 records). The aim is to join areas into clusters so that all of them fulfill conditions to be considered as Labour Market Area. In each step only two areas are joined. Then computations for all new areas need to be performed from the beginning. There are also certain situations when once joined areas are divided, which makes the whole process more complicated. R seems to be a proper tool to carry out necessary analyses of such a big dataset.