I’m a market-oriented economist (is there really any other kind?) so as my Financial Post column about it tried to make clear, there wasn’t a lot I liked in Tuesday’s very market-skeptic federal budget.

Many of the problems the budget promises to solve I don’t think are problems in the first place. (Inequality, for example. See me talk about that with Steve Paikin.) And even if they were problems, the budget doesn’t make a convincing case—it doesn’t even try to—that what it proposes, which often involves hiring more people to study and analyze these problems, will solve them.

To be sure, more study and analysis will directly contribute to the budget’s goal of a “strong middle class:” most of the analysts hired will be well-educated professionals. On the other hand, the money to pay them will come from taxes on the middle class, so it’s not clear there will be a net gain for that class. (If you like class analysis, it would be better to “make the rich pay” but there just aren’t enough rich to pick-up more than a small part of the bill.)

There’s one part of the budget I do like, however. A lot. Pages 108 and 109 of the Budget Plan consist of a table titled “Streamlining the Innovation Program Suite.” It’s printed in a single-digit font, and with good reason. It shows how Ottawa is going to consolidate 92 existing programs designed to encourage innovation into just 35—or, to be completely accurate, “35+.” Pointing to this table, the document declares, in bold type no less, that: “Total overall funding for innovation programming will increase, but the reform will see a reduction in the total number of business innovation programs by up to two-thirds.”

I saw that “up to two-thirds” before I saw the table and thought, Wow, that’s serious, nothing ever gets cut by two-thirds! But then I saw that even after a two-thirds cut—much of which is promised, not actually achieved yet—there will be 35+ innovation programs left. And those are just the federal programs. Provincial and municipal governments all have their own innovation programs, too.

Meditating on the “35+” produces a couple of thoughts. Just about every new innovation program ever introduced has been sold in breathless prose implying that what’s being done is bold, new and, well innovative. (The words “new and innovative” occur five times in the 2018 Budget Plan, “innovative” on its own, 28 times.) Lots of budgets have set-up what were supposed to be new and innovative ways to encourage innovation. That’s why we have 92 programs already. That weak Canadian innovation is still considered a big enough problem that we need even more programs to address it suggests the 92 that have been tried—or at least the 57 that are going to be eliminated to get us down to 35+—haven’t achieved what they were supposed to. If more innovation programs were all we needed to solve our innovation problems, they would have been solved a long time ago.

That gets me to the other part of the budget I really like. On page 107, just before showing us the chart on streamlining, the government says it will provide “$1 million per year… to Statistics Canada to improve performance evaluations for innovation-related programs,” as well as $2 million a year to the Treasury Board Secretariat “to establish a central performance evaluation team to undertake innovation performance evaluations on an ongoing basis, including using the data developed by Statistics Canada."

There’s lots that could go wrong with that.

The evaluations might not ask hard counterfactual questions, such as “How much of the activity supported would have taken place anyway, without government support?” And $3 million a year is not very much money—not in a budget with a net fiscal impact of $23 billion over six years—to perform what should be the crucial function of, as the title of this section of the budget says, “Placing evidence at the centre of program evaluation and design.” A better number might be $10 million or $15 million or even more million.

But given that we have 92 programs to encourage innovation already and yet continue “to face challenges when it comes to translating the ideas generated from [our] world-class research into goods and services that people can use [page 93]”, spending more money evaluating whether programs achieve what they’re supposed to seems a purpose just about everybody should be able to support.