Welsh Journals

Some recent parliamentary boundary changes and the February 1974 general election in Wales A. H. TAYLOR Tutor in Geography University College of Swansea It is well known that the exact geographical location of constituency boundaries has a considerable effect on election results. Changes in constituency boundaries are necessary from time to time due to shifts in population. During this century the main reasons for boundary changes have been the declining populations of city centres and older industrial areas, and the growth of suburbs. Since the areas with declining populations are predominantly Labour (Rowley, 1970) while the growing suburbs are predomin- antly Conservative, one would expect the boundary system to be biased towards Labour shortly before each boundary revision because the Labour seats would have the smaller populations. The boundary changes should put this right. But it is now recog- nized that this is not the case, and the problem is more complex. The effects of a boundary revision Calculations of the number of seats each party could expect to win from the votes it polled show that the boundary system helps the Conservatives just after one of the periodic boundary revisions, and that this bias gradually fades as the next revision approaches. This is why the Conservatives won the 1951 general election in spite of polling fewer votes than Labour. The election came not long after the 1948 boundary revisions. The reason why the pro-Conservative bias declines is that the changes in population mentioned above gradually come to counteract this bias. The reasons for the existence of this bias in the first place concern the statistical distributions of support for the two major parties. Labour has a large number of seats in which it has slightly below average support, and a smaller number of seats such as mining and city centre seats where it has well above average" support. Therefore if it obtains fifty percent'of the votes it can expect to win comfortably