Lijphart seeks to test which type of democratic institutions - consensus or majoritarian - performs most effectively. He tests the performance of these institutions through a statistical analysis of their relative efficiency in three broad fields: macroeconomic management, control of violence, and what he terms the "kinder and gentler" qualities of democracy (293). However, before discussing the results of Lijphart's study, it is necessary to explore what distinguishes the institutions of majoritarian and consensus systems.
Lijphart distinguishes between these two types of democracy by illustrating ten institutional differences which divide the typologies. For clarity, the author divides these ten differences into two distinct dimensions: executives-parties, and federal-unitary. The executives-parties dimension addresses "the arrangement of executive power, the party and electoral systems, and interest groups" (3). The federal-unitary dimension illustrates differences in institutional structure of a federated versus unitary government.
According to the executives-parties dimension, the majoritarian system, or Westminster model, is found to have a two party system and a strong one-party executive and cabinet. Often the executive is more powerful than his or her legislative counterparts. Furthermore, a majoritarian system often uses a single member district electoral system which can lead to disproportional representation, and has a highly competitive pluralist interest group system. Lijphart cites Britain and pre-1996 New Zealand as majoritarian systems.
Lijphart's consensus democracy varies institutionally from the Westminster model. First, under the majoritarian model, the executive office is often composed of a multi-party power-sharing cabinet or coalition. In addition, power-sharing exists between the executive branch and the legislature the electoral system often promotes proportional representation. Lastly, unlike the highly competitive special interest group system of the Westminster model, a consensus democracy promotes a system of interest group compromise (4). Lijphart uses Switzerland and Germany as examples of consensus democracy.
According to the federated-unitary dimension, the Westminster system has a strong, centralized government and a unicameral legislature. In addition, most majoritarian systems possess a very flexible constitution that can readily be amended or changed. Furthermore, in many majoritarian systems, the legislature holds the final word in the constitutionality of legislation, and as such, majoritarian systems do not have a strong system of judicial review.
The consensus model, on the other hand, often has a decentralized government, and can be a federated system. Often the legislature is divided into two houses. In addition, the constitution is often rigid, making change difficult. Lastly, the consensus system often has a strong institution of judicial review to monitor the legality of legislation.
To test the effectiveness of consensus and majoritarian systems, Lijphart compares the performance of the two democracies on three main categories: macroeconomic management, levels of political violence, and the "kinder, gentler" aspects of democracy. Lijphart's hypothesis "is that consensus democracy produces better results - but without the expectation that the differences will be very strong and significant" (261).
When exploring the effectiveness of the two democracies in macroeconomic management, the author operationalizes a number of variables. For the sake of brevity, I will condense the findings into six categories: economic growth, inflation rates, unemployment, strike activity, budget deficits, and freedom index. Lijphart tests the performance of the democracies by using both the executives-parties dimension and federated-unitary dimensions.
In the case of economic growth using the executives-parties dimension, there was little difference between majoritarian and consensus democracy. There was a weak negative relationship between consensus democracy and economic but the findings were not statistically significant. This implies that the difference between consensus and majoritarian democracies in regards to economic development is negligible. In regards to inflation, Lijphart finds that consensus democracies have a slightly lower rate of inflation than majoritarian systems. Consensus also performs slightly better than the majoritarian model in regards to unemployment, but again, the differences are slight.
Interestingly, Lijphart found a massive relationship between strike activity and consensus democracy. According to the regression coefficient, levels of strike activity would have been substantially lower in consensus systems than in majoritarian. However, upon further analysis the relationships are not statistically significant and as Lijphart illustrates, the large difference is a result of "big exceptions to the tendency of consensus countries to be less strike-prone than majoritarian democracies" (269). Lastly, Lijphart explores the performance of consensus democracies on budget deficits and economic freedom. Again, the author finds the differences negligible. When using the federated-unitary dimension, Lijphart's finding are similar except when looking at the inflation variable. When comparing consensus democracy on federal-unitary dimension on inflation, Lijphart discovers that a strong negative relationship exists, the relationship is statistically significant, and there is an acceptable t-value. The author explains this relationship by citing that in a consensus democracy the central bank independence. Lijphart writes, "the most important reason why central banks are made strong and independent is to give them the tools to control inflation" (273).
In conclusion, the author writes, "the evidence with regard to economic growth and economic freedom is mixed, but with regard to all of the other indicators of economic performance, the consensus democracies have a slightly better record and a significantly better record as far as inflation is concerned" (270).
The results regarding the performance of consensus and majoritarian democracies in controlling political violence are also rather vague. Statistically, the consensus system is slightly violent than the majoritarian system. However, Lijphart contends that the significance of the relationship declines when other variables are controlled and outlying observations are removed. Ultimately, Lijphart contends that the statistics show "at least a slightly better performance of the consensus democracies" (271).
The last group of variables that Lijphart addresses is what he terms the "kinder, gentler" aspects of democracy. The author contends that consensus systems are more apt to be "kinder and gentler" than their majoritarian counterparts. Lijphart writes, "Consensus democracies demonstrate these kinder and gentler qualities in the following ways: they are more likely to be welfare states; they have a better record with regard to the protection of the environment; they put fewer people in prison, and are less likely use the death penalty; and the consensus democracies in the developed world are more generous with their economic assistance to the developing nations" (275-6).
Lijphart measures the effectiveness of consensus intuitions by measuring a number of variables: women's representation, political equality, electoral participation, satisfaction with democracy, government-voter proximity, and accountability and corruption. Statistically, Lijphart's findings when comparing the performance of consensus and majoritarian democracies in regards to the "kinder and gentler" qualities are much more revealing. Lijphart finds that consensus democracy "makes a big difference with regard to almost all of the indicators of democratic quality and with regard to all of the kinder and gentler qualities" (300).
To conclude, Lijphart has found that the institutions of consensus democracies perform slightly better than majoritarian institutions in both macroeconomic management and in the prevention of political violence. However, the differences are slim and arguably irrelevant. But, Lijphart did discover that when looking at the "kinder, gentler" aspects of democracy, such as women's rights, incarceration rates and other, consensus democracy performed substantially better.