The Encyclopedia Britannica defines Canada's Liberal party as centrist.
If this is true, then their collapse can be explained by a splintering of the electorate, so fewer Canadians identify with the centre, or by an increase in political competition. Now other parties, such as the Greens, compete for the centrist vote.
While both these stories may contain elements of truth, let me propose an alternative explanation: Canada's Liberal party is a centralist party, and centralism doesn't work in a heterogeneous society.
A centralist party concentrates decision-making in the centre - in the federal government, in Ottawa. Decentralizers, by contrast, delegate decision-making to provinces, municipalities, or individual Canadians.
Many of the policies that define the Liberal approach to government are centralist. For example:
- The Canada Health Act sets parameters for provincial health insurance plans. In order to receive federal funding, provincial plans must to satisfy key criteria, such as universal, comprehensive coverage.
- the Canadian Charter of Rights and Freedoms has empowered the Supreme Court of Canada, increasing the scope of judicial review.
- the Canada Child Tax Benefit replaced a patchwork of provincial social welfare programs with a national income guarantee for children.
In the most recent election campaign, the Liberal's centralist approach was exemplified by a promise of universal child care. Yet universal child care also illustrates the challenges centralists face in a heterogeneous society.
Canadian families come in all shapes and sizes. Some parents are employed full-time, some part-time, some not at all. Some work days, some nights, some weekends. Some want the language of care to be English, some prefer French, others prefer other languages. Some prefer licensed, qualified, regulated care; others prefer care provided by a friend or family member. No single model of care works for all types of families. And this may be one reason why the Liberal party's universal child care proposal failed to gain traction with voters. (Another possible explanation is that an aging electorate isn't interested in paying taxes for other people's children).
There is an old theory called the Tiebout hypothesis, which observes that, if we all have the same tastes and want the same things, then we'll all agree on how governments should spend their money. But if we're different - e.g. one group wants money spent on lawn bowling, another group wants money spent on playgrounds - conflicts arise.
Tiebout's theory suggests that decentralization is a way of resolving conflicts over spending. If people can sort themselves into homogeneous communities - the lawn bowlers in one community, the playground goers in another community - and each community is free to choose their own level of public goods, then public goods can be provided efficiently and everyone can be happy.
Reality is more complicated than this simple story. But there have been a number of studies of the impact of ethnic diversity on the level of government spending, perhaps because this is a measurable form of heterogeneity. One frequently cited study has found that societies with more ethnic fractionalization tend to be more de-centralized. Others have found that ethnic fragmentation tends to be associated with lower levels of government provision of public goods such as education or roads, suggesting that heterogeneity makes it hard to build consensus around government spending.
Other forms of heterogeneity have an impact on spending, although their impact is more debated. Some argue that more income inequality increases the desired level of government spending, others that spending will decrease - the empirical evidence is not clear. Population aging is also pulling up Canada's population pyramid, spreading out the age distribution. Changing media could alter people's values - if Canadians are watching and reading a thousand different forms of media, they will be exposed to a thousand different viewpoints and opinions, which could lead to a greater range of views.
Back in 2005, David Frum offered this description of the Liberal party:
Unlike their supposed analogues, the Democrats in the United States or Great Britain's Labor Party, Canada's Liberals are not a party built around certain policies and principles. They are instead what political scientists call a brokerage party, similar to the old Italian Christian Democrats or India's Congress Party: a political entity without fixed principles or policies that exploits the power of the central state to bribe or bully incompatible constituencies to join together to share the spoils of government.
A bit harsh, but it raises an interesting question:
If the power of the central state declines for fundamental demographic, economic or technological reasons, where does that leave the Liberal party?