State capacity is the measure of a government’s ability to get stuff done that benefits its population. There are many ways to quantify state capacity, including GDP per capita spent on health, education, and infrastructure versus outcomes; the tax-to-GDP ratio; judicial independence; enforcement of contracts; and crime rates.

But a government’s ability to actually implement its own policies has got to rank pretty high here, too. All the best intentions are worthless if, as I wrote in the context of the Liberal’s 2023 national action plan to end gender-based violence, your legislation just won’t work in the real world.

So I thought I’d take a look at some examples of federal legislation from the past ten years that passed through Parliament but, for one reason or another, failed to do its job. We may agree or disagree with goals driving the various initiatives, but government’s failure to get the work done over and over again speaks to a striking lack of state capacity.

The 2018 Cannabis Act (Bill C-45). C-45 legalized recreational cannabis in Canada, with a larger goal of regulating production, distribution, and consumption while reducing illegal markets and protecting public health. However, research has shown that illegal sales persisted post-legalization due to high legal prices and taxation. Studies have also shown continued use among children despite regulations. And there are troubling indicators about the overall impact on public health.

The 2021 Canadian Net-Zero Emissions Accountability Act (Bill C-12). The legislation aimed to ensure Canada achieves net-zero greenhouse gas emissions by 2050 by setting five-year targets and requiring emissions reduction plans. However, critics argue it lacks enforceable mechanisms to guarantee results. A much-delayed progress report highlighted a lack of action and actual emissions reductions lagging far behind projections.

The First Nations Clean Water Act (Bill C-61) was introduced in late 2024 but, as of the recent dissolution of Parliament, not yet passed. This should be seen in the context of the Safe Drinking Water for First Nations Act (2013), which was repealed in 2021 after failing to deliver promised improvements in water quality due to inadequate funding and enforcement. The new bill aimed to address these shortcomings, but a decade and a half of inaction speaks to a special level of public impotence.

The 2019 Impact Assessment Act (Bill C-69). Passed in 2019, this legislation reformed environmental assessment processes for major projects. Many argue it failed to achieve its dual goals of streamlining approvals while enhancing environmental protection. Industry groups claim it created regulatory uncertainty (to put it mildly), while environmental groups argue it hasn’t adequately protected ecosystems. No one seems happy with this one.

The 2019 Firearms Act (Bill C-71). Parts of this firearms legislation were delayed in implementation, particularly the point-of-sale record keeping requirements for non-restricted firearms. Some provisions weren’t fully implemented until years after passage.

The 2013 First Nations Financial Transparency Act. – This legislation, while technically implemented, was not fully enforced after 2015 when the Liberal government stopped penalizing First Nations that didn’t comply with its financial disclosure requirements.

The 2019 National Housing Strategy Act. From the historical perspective of six years of hindsight, the law has manifestly failed to meaningfully address Canada’s housing affordability crisis. Housing prices and homelessness have continued their rise in major urban centers.

The 2019 Indigenous Languages Act (Bill C-91). Many Indigenous advocates have argued the funding and mechanisms have been insufficient to achieve its goal of revitalizing endangered Indigenous languages.

The 2007 Public Servants Disclosure Protection Act (PSDPA). Designed to protect whistleblowers within the federal public service, the PSDPA has been criticized for its ineffectiveness. During its first three years, the Office of the Public Sector Integrity Commissioner (OPSIC) astonishingly reported no findings of wrongdoing or reprisal, despite numerous submissions. A 2017 review by the Standing Committee on Government Operations and Estimates recommended significant reforms, but there’s been no visible progress.

There were, of course, many bills from the past ten years that were fully implemented.¹ But the failure rate is high enough that I’d argue it should be taken into account when measuring our state capacity.

Still, as a friend once noted, there’s a silver lining to all this: the one thing more frightening than an inefficient and ineffective government is an efficient and effective government. So there’s that.

  By David Clinton

Shaping Government Spending Choices to Reflect Taxpayer Preferences

Technically, the word “democracy” means “rule of the people”. But we all know that the ability to throw the bums out every few years is a poor substitute for “rule”. And as I’ve already demonstrated, the last set of bums you sent to Ottawa are 19 times more likely than not to simply vote along party lines. So who they are as individuals barely even matters.

This story isn’t new, and it hasn’t even got a decent villain. But it is about a universal weakness inherent in all modern, nation-scale democracies. After all, complex societies governed by hundreds of thousands of public servants who are responsible for spending trillions of dollars can’t realistically account for millions of individual voices. How could you even meaningfully process so many opinions?

Hang on. It’s 2025. These days, meaningfully processing lots of data is what we do. And the challenge of reliably collecting and administrating those opinions is trivial. I’m not suggesting we descend into some hellish form of governance by opinion poll. But I do wonder why we haven’t tried something that’s far more focused, measured, and verifiable: directed revenue spending.

Self-directed income tax payments? Crazy, no? Except that we’ve been doing it in Ontario for at least 60 years. We (sometimes) get to choose which of five school boards – English public, French public, English separate (Catholic), French separate (Catholic), or Protestant separate (Penetanguishene only) – will receive the education portion of our property tax.

Here’s how it could work. A set amount – perhaps 20 percent of the total federal tax you owe – would be considered discretionary. The T1 tax form could include the names of, say, ten spending programs next to numeric boxes. You would enter the percentage of the total discretionary portion of your income tax that you’d like directed to each program with the total of all ten boxes adding up to 100.

The specific programs made available might change from one year to the next. Some might appear only once every few years. That way, the departments responsible for executing the programs wouldn’t have to deal with unpredictable funding. But what’s more important, governments would have ongoing insights into what their constituents actually wanted them to be doing. If they disagreed, a government could up their game and do a better job explaining their preferences. Or it could just give up and follow the will of their taxpayers.

Since there would only be a limited number of pre-set options available, you wouldn’t have to worry about crackpot suggestions (“Nuke Amurika!”) or even reasoned and well-meaning protest campaigns (“Nuke Ottawa!”) taking over. And since everyone who files a tax form has to participate, you won’t have to worry about a small number of squeaky wheels dominating the public discourse.

Why would any governing party go along with such a plan? Well, they almost certainly won’t if that’s any comfort. Nevertheless, in theory at least, they could gain significant political legitimacy were their program preferences to receive overwhelming public support. And if politicians and civil servants truly believed they toil in the service of the people of Canada, they should be curious about what the people of Canada actually want.

What could go wrong?

Well the complexity involved with adding a new layer of constraints to spending planning can’t be lightly dismissed. And there’s always the risk that activists could learn to game the system by shaping mass movements through manipulative online messaging. The fact that wealthy taxpayers will have a disproportionate impact on spending also shouldn’t be ignored. Although, having said that, I’m not convinced that the voices of high-end taxpayers are less valuable than those of the paid lobbyists and PMO influencers who currently get all the attention.

Those are serious considerations. I’m decidedly less concerned about some other possible objections:

• The risk that taxpayers might demonstrate a preference for short term fixes or glamour projects over important long term wonkish needs (like debt servicing) rings hollow. Couldn’t those words just as easily describe the way many government departments already behave?
• Couldn’t taxpayer choices be influenced by dangerous misinformation campaigns? Allowing for the fact the words “misinformation campaign” make me nervous, that’s certainly possible. But I’m aware of no research demonstrating that, as a class, politicians and civil servants are somehow less susceptible to such influences.
• Won’t such a program allow governments to deflect responsibility for their actions? Hah! I spit in your face in rueful disdain! When was the last time any government official actually took responsibility (or even lost a job) over stupid decisions?
• Won’t restricting access to a large segment of funds make it harder to respond to time-sensitive emergencies? There are already plenty of political and policy-based constraints on emergency spending choices. There’s no reason this program couldn’t be structured intelligently enough to prevent appropriate responses to a genuine emergency.

This idea has no more chance of being applied as some of the crazy zero-tax ideas from my previous post. But things certainly aren’t perfect right now, and throwing some fresh ideas into the mix can’t hurt.