Policy analysis allows provides a structure whereby problems are assessed and solutions are systematically developed and presented[1]. The analytical framework used in this report is a ‘comparative institutional analysis’ of policies from several countries related to aspects of road safety. This kind of approach is useful when we have reason to believe that new policies can be informed through a comparison of approaches taken to address a similar issue in other jurisdictions[2]. This approach therefore allows new policies to be rapidly formulated through a careful analysis of those policies that have proven effective elsewhere, learning from them, and applying them towards local settings[3]. Comparative analysis of this kind also enhances our understanding of contextual elements which shape the theoretical problem of interest, and allow causal inferences to be drawn[4].
For this study, a number of countries from the OECD were chosen for comparative analysis. These countries were selected in order to include a broad range of fatal crash rates, ranging from the Netherlands with 4.3 deaths/ 100 000 population through to the USA with 14.2 deaths /100 000 population annually. Countries outside the OECD were not considered due to the comparative difficulties in obtaining information, and the lesser applicability of the results of these countries to the New Zealand context, due to differences in cultural factors as well as in capacity to implement and enforce policies. A comparative analysis was then performed on each of these countries for a number of variables thought to be correlated with differences in relative road tolls, in the hope that clear patterns would emerge for some of these variables, thereby allowing effective policy recommendations to be drawn from this.
As identified previously, a wide range of factors, both at the individual and wider contextual level, influence the potential of a road accident occurring. Considering each of these in turn is beyond the scope of this study, which therefore focuses solely on individual level factors. This focus was chosen was several reasons, both practical and theoretical. First, it is believed should individual factors prove to be significant, it will be comparatively easy to design and implement policies targeting these, whereas contextual factors such as road design or weather conditions will prove either prohibitively expensive or logistically impossible to change. Second, this report argues that, while contextual factors in themselves add to the risk, these factors require an individual component in order for risk to convert to an actual accident, and that therefore changing these contextual factors are less important than changing how individuals respond to them. For example, while wet weather increases the risk of a crash occurring, the more effective strategy for reducing this risk is likely to be improving driver awareness and encouraging reduced speed in wet conditions, in other words, changing individual behaviours.
As mentioned previously, young drivers are disproportionally represented in fatal crash statistics. It is also worth noting that within countries in the OECD, a close correlation exists between high overall fatality rates, and high fatality rates for people asked between 15 and 24 (refer to figure 2). In other words, countries with low road tolls tended to have low road tolls for young drivers, while the opposite was true in countries with high road tolls. For this reason, the factors considered in this study were further narrowed down to enable policies targeting young drivers to be evaluated.
| Country | Deaths/100 000 population | Alcohol limit % | Minimum licensing age | Number driving hours required to get license | Speed limit | ||||
| Adult drivers | Young drivers | Provisional/learners | Full | Motorway | Open road | Urban | |||
| Netherlands | 4.3 | 0.05 | 0.02 | None | 18 | 0 | 120 | 80 | 50 |
| UK | 5 | 0.08 | 0.08 | 17 | 17 | 0 | 110 | 100 | 50 |
| Norway | 5 | 0.02 | 0.02 | 16 | 18 | 35 | 90 | 80 | 50 |
| Germany | 6 | 0.05 | 0 | 17 | 18 | 30 | None | 100 | 50 |
| France | 7.5 | 0.05 | 0.05 | 16 | 18 | 20 | 130 | 90 | 50 |
| Australia | 7.7 | 0.05 | 0.02 | 16 | 18 | 50 | 10 | 100 | 50 |
| Ireland | 7.8 | 0.08 | 0.08 | 17 | 17 | 120 | 80-100 | 30-60 | 60 |
| Canada | 9.2 | 0.08 | 0 | 16 | 18 | ||||
| New Zealand | 10 | 0.08 | 0.03 | 15 | 16 | 0 | 100 | 100 | 50 |
| Belgium | 10.1 | 0.05 | 0.05 | 17 | 18 | 20 | 90 | 120 | 50 |
| USA | 14.2 | 0.08 | 0.08 | 14-16* | 16-21* | 0-60** | 110 | 80 | 60 |
[1] David L. Weimer & Aian R. Vining (1999) Policy Analysis: Concepts and Practice (New Jersey: Prentice Hall, 1999), 39
[2] Mintrom, Michael. (forthcoming 2011). Contemporary Policy Analysis. New York: Oxford University Press, 241
[3] Mintrom, Michael. (forthcoming 2011). Contemporary Policy Analysis. New York: Oxford University Press, 242
[4] Peter Burnham, Karin Gilland, Wyn-Grant, and Zig Layton-Henry, Research Methods in Politics (New York: Palgrave Macmillan, 2004), 57.