Tax policy and open government

Tuesday, September 19, 2017

by Kent Aitken RSS / cpsrenewalFacebook / cpsrenewalLinkedIn / Kent Aitkentwitter / kentdaitkengovloop / KentAitken

The Government of Canada has proposed changes to the tax framework, which has struck a wide variety of nerves. Moving swiftly past any discussion of that policy proposal itself, it provides a solid example of the environment in which government works today. I still get interest in explanations of what the idea of open government actually means, and this lets us move from the abstract to the practical.

It’s worth noting that while this public debate has emerged recently, the proposal comes from the 2017 federal budget, including some of the tax strategies the government was examining. Any policy discussion exists in the context of the government’s agenda, which shows up in party platforms, budgets, and the Speech from the Throne. And that agenda, of course, came from somewhere in turn: long-standing party positions, industry lobbying, think tank research, government advice, concerns of MPs’ constituents, international agreements, etc. There’s a complex ecosystem of inputs.

The Department of Finance started inviting comments on July 18. Far from the lightweight budget consultations of years past (essentially a webform and the question “What should the government prioritize?”), they posted a brobdingnagian backgrounder with rationale, data, and their calculations. If they posted the raw datasets, I couldn’t find them; the open government ideal would be linking to the same data that government analysts are using (i.e., on Economists, researchers, developers and transparency advocates have long been scraping or copy and pasting information from government websites, which is a barrier to use.

Finance has asked that people read the 27,000+ word backgrounder and provide written comments. There are design decisions baked into this.

One, it’s an email inbox, not an online discussion forum. The written comments aren’t automatically publicly visible, and organizations can consent to publishing or stay mum for privacy.

Two, while the consultation and backgrounder is public, their target audience is clearly a limited group of experts, academics, and businesses (most likely industry organizations that are funded and mandated to analyze and respond to such proposals). The most visible government consultations in recent years have been questions like Ontario Budget Talks, federal electoral reform, or the Innovation Agenda. Those are geared towards broad audiences. However, most government consultations, running quietly for decades under the radar, are closer to this Finance example. The standard has been that government posts an analysis or a draft regulatory change and asks a small community of experts for detailed comments (e.g., on wildlife management strategies or chemical allowances in products). Anyone could respond. But most wouldn't have any fun doing so.

There’s no right or wrong on these two items, just different approaches for different contexts. Making every comment publicly visible can be disenfranchising to many would-be participants. And governments should (generally) design and promote for their actual target audience of stakeholders.

For most people, including many small business owners, their engagement with this consultation will not be in reading the backgrounder and commenting. Instead, they’ll read analyses, op-eds, and updates in the media. Others may contact their MP with their concerns. Others will continue to pay dues to lobbying organizations to do this work on their behalf and trust that their views are represented. This reality will be true of most examples of “open government;” most engagement is through “info-mediaries” rather than direct contact.

The next important feature: government can strategize and plan for the public reaction, but once in the wild, it’s to a large degree out of government’s control. There has been a flurry of media articles written since July 18, taking many different tacks (e.g., that it’s unfair and disincentivizing to doctors specifically) (again, no value judgment here).

For the potential value for government, let’s use Dr. Kevin Milligan as an example. The start of his Twitter thread on the proposal is probably the most-liked tweet ever that starts with “I’ve made spreadsheets…”

Milligan, an Economics professor at UBC, reworked and re-calculated a lot of the data from Finance (and, admirably, posted his work and spreadsheets publicly for government officials or other observers to review). His analysis supports much of Finance’s work, calls some parts into question, but definitely adds value. This is the entire point of of open government and open data: recognizing that expertise exists outside government’s walls and creating ways to work it into public decision-making. Milligan’s analysis, alongside a number of other prominent Canadian economists who contribute actively to public debate, turned into blogs or columns in print and online media. That community, plus pundits and other observers, also debated parts of the proposal and each other’s analyses in short form on Twitter.

The email inbox is still central - Milligan encouraged people to write into it - but it’s one of many ways that people will try to influence the decision, even during a formal consultation with an “official” avenue for input.

Milligan’s work (here’s his blog post) is not far off what we’d trendily call “civic tech” (admittedly it’s relatively low-tech civic tech)*. The more common examples are people creating tech platforms for others to find and understand data and information (e.g., or But the point isn’t the platform, it’s the analysis and the value in the context of a public decision. So Milligan’s analysis and spreadsheets meet that goal and that description for me.

Lastly on the timeline, there’s one big step left: the government will decide what it wants to do in Cabinet. The proposal as originally written by Finance will almost certainly still be the starting point, but the public reaction and expert comments will change or at least flavour that. While the final advice to Cabinet and the discussion is held in confidence, Access to Information laws, modern government communications, public engagement, and social media make that a more limited concept than it was in the past. The black box is getting smaller and smaller. But the point remains: even if the elected government talks to everyone about a proposed policy, it’s still on them to make the decision and to be held accountable for it. And they do this out of the public eye, by design.

So having walked through the example, what’s left? A couple closing thoughts:

One, a common theme is that as open government and telecommunications technology make the world horizontal and connected, government policy analysts will be less subject matter experts and more facilitators and convenors of stakeholders and external experts. I think this take is wrong, and Finance provides the counterexample. They started with a 27,000-word backgrounder, and I’d bet that there’s a much longer version behind it. They need to be able to understand, fact-check, and contextualize technical input from experts. They need to know if Milligan’s numbers are right or wrong when they lead to game-changing conclusions. So I’d actually propose that government analysts have to be better than they’ve ever been to navigate the modern governance environment.

Two, open government can’t be about changing governance models. How government does things always has to serve the things government needs to do. Open government is better found, in the long run, in the nuts and bolts of government policies and programs. In this case, in the data, principles, and analysis behind tax policy.

Alignment and Competition

Friday, September 1, 2017

by Gray O'Byrne RSS / cpsrenewalFacebook / cpsrenewalLinkedIn / Gray O'Byrnetwitter / GrayOByrne

“Let’s set up a meeting to make sure our initiatives are aligned and we’re not duplicating effort.”

I have become wary of this line. On the surface, I couldn’t agree more. We want our time to be well spent. The phrase taps into my human aversion to waste, but like everything else, it can be taken too far.

First off, duplication is sometimes necessary. In a sense it’s impossible to replace an existing system without duplicating it. As we build digital services for citizens, we are duplicating the existing ones. Introducing digital services may not be controversial but what if the existing system is trying to slowly digitize itself? How do we decide if it’s worthwhile trying something new rather than invest more in the improvement efforts?

In a lot of cases there is no clear answer and rather than investing in the new idea, we align ourselves with the existing initiative to avoid duplication. This could be contributing to how long it is taking governments to built digital-first services.

When developing new services we also have a choice to make about what tools or approaches to use. Sometimes a single team has the resources and expertise to try out several options instead of picking just one, but what happens when it’s two different teams each with their own idea? We seem to have very little tolerance for allowing both teams to continue even if it’s unclear which idea is more promising.

In this scenario, instead of waiting to see what we learn from each team, we tend to ask one of them to stop. I’ve seen several promising initiatives shut down this way. The project team eventually met with a group who had more authority and who felt their own project would be doing the same thing. “Why would we create two solutions to the same problem?”

These tendencies to stop others from trying different approaches to solving the same problem are what I am pushing back against. We need to be investing in a variety of approaches because they can all teach us something. Sometimes the alternative projects will result in better services, sometimes in wasted effort but if done properly we will always learn.

Unfortunately multiple groups tackling the same problem means there will be competition between ideas, projects and even the people who want to see their own approach succeed. But there are benefits to competition as well! We’ve come to accept that it is a net positive and helps drive innovation In the private sector and just because there is some competition between groups, it doesn’t mean we need to have winners and losers vying for space in government.

Rather than aligning on execution we can align our goals. Through collaboration we can ensure each project takes a unique approach so it will generate unique insights. Information sharing at all stages will be key, we need to experiment and learn as a whole. The hard part will be for everyone to come together once individual projects have been tested, but it can be done.

I’ve had some success with this approach. 

When collaborating to build a micro-tasking platform for the federal government, my team took a conscious step back from the working group. We knew we could rapidly deploy something in our organisation and felt the group could learn more from us building an advanced prototype than if we participated in the same way as everyone else.

We kept in touch, we built in the open and challenged our colleagues to make something better. This was deliberately duplicative but we got to benefit from our prototype quickly and were able share our work and insights back with the group. A year later when the working group had their solution in place, it had many of the features we had designed. At this point we shut our prototype down so our users would join the superior platform and benefit from the broader community that had access to it.

The strength and weakness of this approach are the same to me: we are working with humans. On one side, a little competition and a strong sense of ownership are both extremely motivating. Trying out a few different ideas in earnest can also help when our intuitions are wrong. On the downside, competition is uncomfortable. If people feel like “it’s us or them” they could stop sharing openly and even create barriers for others. As a believer in human awesomeness, I think these downsides can be managed by aligning our goals (not our approaches) and collaborating openly within our organisations. To get us started though, we will need to become comfortable with the idea that attacking a problem from multiple angles is an acceptable form of duplication.