Bad Senate ideas: provincial appointment

(This may or may not turn out to be a regular series, depending on how long the current infatuation with Senate ponderings lasts.)

So Canadians are talking about reforming or even (gasp!) abolishing the Senate. Abolishing the Senate is constitutionally difficult, but analytically pretty simple: we wouldn’t have a Senate anymore, and the House of Commons and Crown would constitute the Canadian Parliament[1].

With either abolition or reform, we have to ask what the effects would be, and whether the means would serve the intended ends.

So in that spirit (and because I just read an interesting paper about it) let’s all agree that as bad as having the Prime Minister appoint Senators is, having the provinces appoint them would be worse.

“What the eff?” you say. “Who’s even advocating that?” Well, nobody important that I can see at the moment. (Though parts of this Tom Flanagan piece could be read that way.) But this was a common argument by provinces in the 1970s, and god knows bad ideas never really go away.

And it’s a bad idea not because I don’t like it, but because it fails to achieve its stated goal: increasing sub-national control over the national government. In fact, history shows it does the opposite.

Before 1913, the US Senate was appointed by state legislatures. This was explicitly intended to give the states control over the President, and not incidentally the Supreme Court. Then, after decades of campaigning, in just 11 months the Congress and two-thirds of the states passed the 17th amendment to require directly elected Senators.

A new paper from David Schleicher of George Mason University explains why states voted to disempower themselves. Schleicher argues that the role of national political parties in the US made the Framers’ conception of federalism unworkable, and in fact put state politics at the mercy of national politics–the exact opposite of what the founders intended.

Why? It’s pretty simple: state elections became “nationalized”, with voters expected to elect state representatives based on which party’s senators they would send to Washington. This had already begun before the Civil War, but by the 1870s through the 1900s it became more and more intense. Schleicher cites an 1894 editorial in the Chicago Tribune which explicitly argued that it was wrong for the state’s voters to make their decision based on state issues–national politics was what mattered:

Do these Democratic State Senators think the voters can be called off from the national issues involved in the direct election of Representatives and the indirect election of a Senator to consider only local questions. That they will drop the Wilson bill and devote their attention to the establishment of a Police Board in Chicago? That they will lose their interest in the currency — in the silver question and the taxation of State bank notes–and become wrapped up in the question whether the Chicago park boards shall be elective or appointed?

The 13th amendment, then, was a way of restoring the spirit of the federalist constitution by abandoning the letter.

The same issue faces other appointed upper houses. The German Bundesrat is an extreme example of this, where the German states (Lands) send delegates who have to strictly follow their government’s instructions. The intention was to decentralize the German government after World War II. The result has been to “nationalize” Land elections, as it was in the US senate. From Bruce Ackerman, 2001 (PDF):

The voters do not independently elect members of the Bundesrat. Its members are representatives of each Land government and strictly follow its instructions. This means that voters in Land elections cannot concern themselves only with the competing parties’ performance at the Land level. They must also bear in mind that their votes in Land elections can shift the balance of national power by changing the party balance in the Bundesrat…

The result has been the nationalization of state politics. National politicians and parties cannot look upon the fate of state elections with relative indifference. They make them part of the national political game, seeking to transform state elections into votes of confidence on the Chancellor and his initiatives. Voters in state elections do not focus only on the promises and performance of their state governments. They tend to use their votes to send a message to Berlin about their satisfaction with the ruling coali-tion on the national level.

There’s some important broader lessons about reform here. The biggest would be that the law of unintended consequences still applies: structures that are nominally intended to preserve local political importance can, in fact, diminish it.

But I want to rest on one point from Schleicher that I think is broadly relevant: the need to understand how we structure institutions with an understanding of how actors will work within them. The lessons of both the US Senate and Bundesrat suggest, for example, that an effort to de-politicize the Canadian Senate by appointing “non-partisan” technocrats will be short-lived at best. It’s a legislature: it’s politics all the way down. National political parties are here to stay, and thinking that they’ll somehow stay out of the Senate if we tinker with some rules is just juvenile.

[1] Fixed from an earlier version where I erroneously wrote that the Commons would be the Parliament alone.

2 thoughts on “Bad Senate ideas: provincial appointment

  1. I still think we need “Second House” to review items passed by the Elected House of Commons, but need to remove politics and parties for the equation.

    How about Senate as “Jury Duty”, with a Single 10 year term – and then vetting of “random” candidate-Senators pulled from the Electoral Rolls of the regions…?

    Yes, the Jury system is NOT perfect – but neither is the current-model.

  2. The problem with voting for a political party is you functionally stop having your MP, whose job it is to help you when you’re having a problem with the machinery of government you can’t resolve using the regular channels. This is pretty much inevitable when you’re looking a party-vote system, and it only gets more pronounced when you’re looking at any kind of proportional representation system. But political party votes are an obvious way to handle complexity and they’re not going away, so, hey, the Senate’s certainly not useful to the Dominion at large right now, we can use it as a fix for this.

    If the Commons is the house of political parties, then the Senate (could be) the house of geographic representation.

    Divide Canada into areas, ignoring provincial boundaries, since this is a federal office. The areas have a population of about a million people, but using a formula that counts land area as well, so sparsely populated northern regions aren’t ignored. If the area has a net inflow of taxes, it’s a region, with two senators; if it has a net outflow of taxes, it’s a city, with three.

    Senators serve a fifteen year term; the first bunch draw straws to see who is up for re-election in five and ten years. Regular, every-five-year elections. You have to live in the area you represent, you can’t have ever been (or become, once elected) a member of a board of directors, a member of a political party, or have run for federal or provincial elected office. Use single-transferable-vote; we’re trying to elect the person least objectionable to the area as a whole.

    Every ten years, areas are evaluated for city-vs-region. (So the net-flow-of-taxes formula has to be written down in law somewhere.) If a senator is lost, the senior senator is lost. If a senator is gained, they’re gained at the next election. (so we’d better co-ordinate the 10 year evaluations with the senate elections.)

    And yes, it’d open the whole constitutional can of worms. But if that elected Senate had to agree to pass legislation, I think it’d have a beneficial effect.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>