Trends in the number of fatalities, vehicles and also population over the period 1985/86 to 1995/96 are given in Appendix B.
Figures 4.1, 4.2, 4.3 to 4.4. show the year to year values of deaths in the different countries by region and it can be seen that only in East African countries are data reasonably complete and showing some of the lowest percentage increase in road deaths over time. From Appendix B it can be seen that there is a very large variation in the change in fatalities over the given ten year period (for those countries where trend data were available). Thus in Uganda there was over a 200 per cent increase in fatalities with Malawi, Benin and Botswana also showing increases of over 100 per cent. Conversely some countries show decreases in deaths from 1985/86 to 1995/96 ranging from two percent in Zambia to over 30 percent in Nigeria.
The decrease in deaths in Nigeria over recent years was commented upon in the previous section. Such decreases in road data are possible, for example in the UK where deaths dropped from 6,000 to 3,620 over the period 1980 to 1995. However the methods used to bring about such decreases are well documented and the expenditure in life-saving activities was very considerable indeed. It is difficult to determine whether or not real reductions in road deaths have actually taken place in Nigeria, given the recent economic decline, or whether data recording methods have changed in any way. A detailed investigation of road safety activities in Nigeria is needed before such questions can be answered.
South Africa presents a different situation where road deaths have increased by 5 per cent over the given time period. Efforts made to reduce road crashes in South Africa are again well documented and a 5 percent increase must be set against a 30 percent increase in vehicles over the same period. It is this sort of result that makes the figures presented for Nigeria somewhat suspect.
The overall average increase in road traffic deaths from 1985/86 to 1995/96 in all countries for which trend data were available was found to be 7 per cent. However this value is fairly meaningless from a statistical point of view because the overall value is biased by changes in the two countries of South Africa and Nigeria which have significantly more deaths than the other countries. If these two countries are removed from the analysis then the percentage increase in the other 17 countries where trend data were available was 42 per cent. Those countries showing the greatest increases were mainly in east Africa and south-east Africa and those with the biggest decreases were mainly in west Africa.
Whilst an increase of about 40% over ten years in people killed in road crashes is certainly significant, it is certainly much lower than previous trends found for African countries. Thus in an earlier TRL study (Ghee, Astrop et al 1995) it was found that over the period 1968 to 1983 road deaths in Africa increased by over 250 percent. Caution is needed in comparing this early result with the current study because the members of countries included in the two analyses were not the same. However there is some evidence that very large increases in road deaths over time are now lessening.
This is clearly illustrated in the two major countries of Nigeria and South Africa. As stated from 1985/86 to 1995/6 Nigeria shows a (suprising) decrease in deaths of over 30 percent. Yet from 1960 to 1975 deaths there increased by 400 percent and from 1960 to 1980 by a remarkable 700 percent. In South Africa where in this study the increase over 1985/86 to 1995/96 was 5 percent, earlier statistics show an increase from mid 1960's to mid 1980's of about 120 cent.
Trends over the period 1985/86 to 1995/96 for 14 countries and for shorter time periods for 4 countries are shown in Appendix B. Figure 4.5, 4.6, 4.7, 4.8 show trends over the more recent period of 1992 to 1996 by region. It can be seen that all countries for which data were available showed increases in licensed vehicles. Those with the greatest increase from 1985/86 to 1995/96 were Zambia (about 340 percent) followed by Uganda (almost 220 percent) , Togo (170 percent) and Benin (140 percent). The smallest increases occurred in Senegal (13 percent) and Central African Republic (15 percent).
Zimbabwe illustrates the relationship that exists in any country between economic development and growth in vehicle ownership. For example in Zimbabwe, growth over the given ten year period was amongst the lowest at 26 percent. Over the last ten years the economy of Zimbabwe has had problems and this has clearly affected growth in vehicle ownership.
Trends over the shorter time period 1992 to 1996 are shown, by region in Figures 4.5 - 4.8. It can be seen that highest growth occurred in Uganda (140 percent), Botswana (120 percent) and Burundi (85 percent). (The trend in Togo (not shown) appears peculiar in that from 1992 to 1993 it dropped by 40 percent then increased back to the 1992 level by 1994 then rose dramatically by 1995 by over 200 percent).
From Figures 4.5 - 4.8 it would appear that there is as much variation within regions as between them. Countries showing the smallest increases of under 20 percent included Kenya, Tanzania, Niger, Congo , Zimbabwe and Swaziland. Sierra Leone actually showed a decrease from 1992 to 1996, attributable no doubt to the civil war taking place in that country.
Some countries, such as Uganda, Benin and Botswana which showed large increases in road deaths over time also showed significant increases in vehicle ownership levels. However some countries such as Zambia and Togo showed relatively small increases in fatalities (Zambia actually showing a decrease) whilst at the same time showing large increases in vehicle ownership levels.
As stated above, changes in vehicle ownership levels in developing countries are closely correlated with economic development. This is not necessarily the case in developed countries where levels of ownership are already high. Thus in western Europe for example where the economy has been sound in recent years, the number of licensed vehicles per head of population has changed very little. This is, of course due to the fact that levels were already high and approaching 'saturation ' level. In Western Europe, vehicle ownership levels in the 1990's were already in excess of 500-600 vehicles per 1,000 population in most countries. Conversely, the highest vehicle ownership level found in the African countries studied was 120 (South Africa), whilst in most it was in the range 10-20 per 1,000 population. Thus for many African countries the potential exists for continuous rapid growth in vehicle ownership, accompanied by the inevitable rise in road deaths and injury. It is essential therefore, at the current level of social and economic development, that the countries of Africa begin to increase their level of investment in crash prevention and reduction measures.
Finally Appendix B provides information of population growth rates and it can be seen that between 1985/86 and 1995/96 the population of some African countries was very high indeed. For example in South Africa, Malawi, Cote d'Ivoire and Rwanda population increases were at or in excess of 50 percent. The fact that in some countries growth over that period was 'only' 25-30 percent probably reflected the rapid growth that took place in earlier years. Kenya and Nigeria would be examples of this.
In many developing countries there is a shift from rural to urban areas. For example the population in Lagos, Nigeria was estimated to be 10 million in 1995. UN forecasts are that by the year 2015 it will grow to almost 25 million which would make it the world's third largest city. This makes the potential for vehicle ownership (per head) even greater in future years and from a road safety point of view means that growth in urban deaths and injury might grow at an even faster rate than national trends might suggest.