I very much appreciate Erik Moberg's lengthy, thorough, and painstaking review of my latest book Patterns of Democracy. I also appreciate the opportunity he offers me to respond and to have my response posted on his website. I am happy to accept his invitation and I would like to make the following comments:

1. I had two goals: first, analyzing the different forms that democracies can assume (an initial attempt of which was already included in my 1984 book Democracies) and, second, the performance of these different types of democracy, that is, how well does their decision-making operate, especially with regard to macro-economic policy, and how well do they perform in terms of democratic quality. That is why the subject of performance, after being previewed in the Preface and the first chapter, is not discussed again until chapter 14. I see the majoritarian-consensus contrast as a question of forms of government rather than their performance. But this appears to be merely a terminological disagreement between Moberg and me.

2. Are the differences that I find with regard to the performance ("so what") variables small? I agree that this is the case with regard to macro-econommic policy outcomes. But this is also exactly what I expected to find, and it is a significant finding since multitudes of political scientists have argued in the past that majoritarian democracy works much better in these respects than consensus democracy. With regard to just about all of the democratic-quality variables (and also with regard to what I call the "kinder, gentler" attributes of consensus democracy) the differences are large enough to be statistically significant; they are in the expected direction (favoring consensus democracy); and they tend to increase instead of decrease in magnitude when other potential explanatory variables are controlled for and extreme outliers are removed from the analysis.

3. Moberg's definitions of "logic" and "logical deduction" are clearly stricter than mine. (Incidentally, I tend to use the term "derivation" instead of "deduction.") All that I have in mind is that the most important differences between majoritarian and consensus democracy can be described in terms of the majoritarian-consensus contrast. For instance, the first and most important difference between electoral systems is between plurality and other majoritarian systems on one hand and proportional methods on the other; this difference has an obvious relationship to majoritarianism vs. consensus. To use another example, the most important distinction between cabinets that one finds in the extensive literature on cabinet coalitions and cabinet durability is the difference between minimal winning and oversized cabinets; this again is a difference that can easily be interpreted in terms of the majoritarian-consensus distinction.

4. Moberg questions whether "consensus" is the proper ("fitting") term for the type of democracy that is the opposite of majoritarian democracy. I think that it fits the character of consensus democracy on the executives-parties dimension quite well, but I agree that it does not express what I call the "federalist" characteristics equally well. Perhaps I should have used the separate "consensus" and "federalist" terms. Karl Kaiser has suggested that a better general term would be "negotiation" democracy (see p. 2 of Patterns of Democracy); "bargaining" democracy would be another possibility. I leave it up to other scholars whether they want to change my terminology. I think that it will be confusing instead of helpful if I now switched to different terms from the ones that I used in Patterns of Democracy.

5. On my finding that there are two separate clusters of majoritarian vs. consensus characteristics: I think that it was a plausible initial hypothesis that all ten would tend to occur together. The assumption that democracies that use non-majoritarian procedures would want to use both the shared-power and the divided-power forms seems entirely reasonable to me. The evidence, however, clearly shows that this is not the case, and a good explanation for this finding is, I think, that non-majoritarian democracies tend to have clear preferences for either the shared-power approach or the divided-power approach. One puzzle that is not solved by this explanation is why the two dimensions are almost completely unrelated to each other: that is to say, it makes sense to think of the two approaches as different, but why are they so clearly alternative approaches? I have no good answer to this question.

6. Moberg's definition of a minimal winning coalition is obviously much broader than mine; for instance, a cabinet that contains a party that is not necessary for the cabinet to have at least 50 percent support in the legislature, is always considered oversized by me but, if this party is included for insurance under conditions of uncertainty, a minimal winning cabinet by Moberg. One practical problem with Moberg's definition is that in practice it is very difficult to determine what exactly the reason for the inclusion of the unnecessary party is. Let me also point out that my definition is in agreement with the standard usage in the vast literature on cabinet coalitions (reviewed in Chapter 6 of Patterns of Democracy).

7. Finally, the arguments that Moberg advances for the thesis that in two-party systems the one-party government will be closer to the median-voter than coalition cabinets in multiparty systems is entirely plausible. And it is reasonable to extend this argument to majoritarian systems (which are often two-party systems with one-party cabinets) generally. The problem is that it is not true in practice. I cite and use John Huber's and Bingham Powell's findings on pp. 287-88 of Patterns of Democracy; these show that it is in fact the other way around: in consensus systems, the government is actually closer to the median voter than in majoritarian systems. More about this can be found in Powell's recent book Elections as Instruments of Democracy (Yale University Press, 2000). One explanation is that, while it is true that two-party systems tend to be center-oriented, plurality and other majoritarian election methods tend to have a lot of "accidents" and often fail to produce pure or almost pure two-party systems.