Results and Delivery: Lessons for Canada (with Sir Michael Barber)

Results and Delivery: Lessons for Canada (with Sir Michael Barber)

Description: Sir Michael Barber, renowned authority on government service delivery, shares a methodology that could be adapted in Canada.

Date: May 26, 2016

Duration: 00:10:12

Resolution: 1080p


Matthew Mendelsohn
Deputy Secretary to the Cabinet (Results and Delivery)
Privy Council Office

Sir Michael Barber, thank you for joining us. Welcome here to Ottawa. You set up the original Results and Delivery Unit under Tony Blair's government. Can you describe a bit the motivation and impetus for that, and how that differed from previous approaches in British governments?

Sir Michael Barber
British Educationist
Chief Education Advisor to Pearson and Managing Partner of Delivery Associates

The original delivery unit was set up by the Blair administration at the beginning of his second term based fundamentally on the insight from the Prime Minister and a number of people who had worked with him that in their first term they had promised a lot, made some great speeches, the economy had grown, on the whole people liked the tone and the direction that Blair was setting for the country—but during the election campaign in 2001, the citizens were basically saying to Tony Blair: “Well, we love the speeches and we like the economic growth but you keep promising the health services are going to get better and it hasn't changed," etc., etc. So the impetus for the delivery was to take an agenda for the second term of domestic policy reforms and really deliver them—not just get the laws passed and the white papers written, but actually change the daily experience of citizens, whether it was a child going to school or a patient going to the health service or ending crime on the streets or the quality of the public transport. So the insight was, this time we really have to get it right and make sure we deliver right out, so that the citizen benefits, feels, sees, feels and notices the difference.

As you set up the initial unit, what were some of the big challenges you faced with the public service?

I'd say that there were two or three challenges with getting started out. One with the politicians, which—they had understood they needed to change but getting them to actually change their practice of responding to crises in the media needed something, and getting them to focus on building routines, getting the way—you know, the very basic level, getting the Prime Minister's diary so that it organized his routine stocktakes right through a year. There was one set of challenges there.

With the public service, there were a couple of challenges. One was, most departments were not enthused about having a new unit at the centre of government. They thought, these new units come and go, they're mostly annoying, they mostly generate a whole lot of bureaucracy…we'll see it off if we can. The challenge at the beginning was building the right relationships with the ministers who… We were saying, we will help solve your problems. We're not Tony's spies. We are here to help you get the job done that the Cabinet has agreed. With the top civil servants who were saying, “Well, who are you? Why should we take any notice?" We were saying to them, “Well, we're going to be a small group of people. We're going to keep the Prime Minister focused on this agenda so he won't be changing his mind. We're also going to help you solve your problems, and when you do get them solved we'll give you the credit." So we went through those key relationships and worked out how to get the relationship so that it was functional and that they could see mutual benefit. Then I think crucially after that was getting the data systems in place, which was work that the Prime Minister wouldn't notice but was actually fundamentally important. So making sure that the 43 police forces in England had a similar way of counting crime and reported that monthly. It took about a year, it was absolutely basic to get this agenda done. So I would say those were the main challenges: change in the way the politicians worked, building the relationships with the public service and then getting data systems in place.

What about the integration of the policy development process and the implementation (because sometimes we think of those as two separate phases)? How did you go about integrating those, and was that a challenge as well?

It was a challenge and it did need some work. The key to it was a combination of over time, putting the data system in place, and secondly, having these routine checkpoints (we called them stocktakes), where every two months Blair was meeting, for example, the Secretary of State for Health or the Secretary of State for Education and reviewing progress towards the goals that had been set. Those review meetings in effect became policy development meetings and implementation meetings. So you get a plan that's good enough, you start implementing, you see what happens, the stocktake comes up, you say this is working, and this isn't working so well or it's working in these places and not in those places, and then you adjust and refine the policy as you go. One of the critiques of British government when it fails is that in those cases there was—I'm quoting some political scientist here—a deficit of deliberation. I think what we put in place was serious deliberation, evidence-informed debate, between the Prime Minister, minister and their advisors. Then, a decision. So the policy got continuously refined. A mistake we made at the beginning of the delivery unit, that I think can be avoided here and elsewhere, is you don't need to get delivery plans perfect at the beginning. You just need to get them good enough, get started, and then once you've got your routines in place those plans can be reviewed, refined and improved as you go, and that does integrate policy and implementation.

What about the data and indicators though? Is it crucial to get those right at the beginning or is there a risk that you kind of go through a process and you realize that you can't actually measure or report on what you're trying to achieve?

I think you need to get the best you can to get started, but you will find sometimes that the data systems that you've got are not good enough and you need to refine them over time. One thing is, some of the goals that get set, climate change would be a good example in Canada or in other parts of the world, is you set a goal for emissions in 2030. That's a long time off. How would you know in 2018 or 2019 whether you're making progress towards that 2030 goal? It may not just be on the overall emissions metric. There may be other things that you need to see that you've put in place. So for example, I did some work with some big American public university systems. They're trying to improve their graduation rates, their six-year graduation rate. But if you're a student in semester one of year one, what do we know about what happens in semester one, year one that will indicate whether that student is on track to graduate in six years or not? We discovered that if that student goes to the library, gets trained in how to use the library and starts using the library in the first month, that's a good indicator they're on track to stay. If they don't do it, it's an indicator that they're on track to drop out. That is something you can do something about. So I think in these areas where you've got a long-term goal, working out what the lead indicators are—what are the things that you can do something about in the short term—that will help you. I'll give you another example from Punjab, Pakistan. We're trying to reduce infant mortality. We know that one thing associated with infant mortality in the long run is whether there's a skilled birth attendant present when an expectant mother gives birth. That's something we can do something about. We are making progress. So that tells us—we don't yet know what's going to happen to infant mortality, but it tells us we're probably on track to reducing infant mortality. So I think thinking through the data system not just in the headline goal but the lead indicators that will tell you whether you're on track is really important, and then only with those can you refine the policy.

The UK has a different system of government than Canada, and we live here in a federal system where provinces and territories have many of the levers. Sometimes the federal government has few policy levers on some of the things that we want to impact. How do you think a delivery approach works in a federal system like Canada's?

Well, the first thing I would say about this whole question, which is obviously fundamental to Canadian history and the Canadian present relationships between the federal government and the provincial government… The first thing I'd say is that deliverology or this system will really help because it makes the whole process explicit. So if you've got a climate change agenda or an infrastructure agenda, by setting some goals, by working out what the delivery plan is, you make explicit what it is you're attempting to do and it makes clearer and sharper where you depend on the provinces or the cities or municipalities for delivery. So it will make the problem or the challenge clearer. That's point one.

The second thing is, while Britain—or at least England, anyway—is more of a kind of unitary system than you have in Canada, actually you do need to think through, even in a unitary system, how you're going to engage people at every level of the system, whether it's teachers, school principals and school boards in the school system, or whether it's provinces and the federal government—municipalities, provinces and the federal government collaborating on changing infrastructure, transport, housing or whatever it might be. So I think the deliverology approach will help make it clearer, but in the end it will depend on the quality of the dialogue between the federal government and the provincial government, and over and above that the appeal the federal government makes to the people of Canada, and the extent to which the people of Canada are saying to their provincial governments as well as the federal government, “Yes, this is what we want to do, we do want to see that happen. We would like Indigenous people to have better opportunities and close that gap. We would like to have better public transport in our big cities. We do want it delivered as soon as possible. So we want you to solve these problems and we don't want you to turn it into one of the many stand offs in Canadian history between the provinces and the federal government."

So I think there's two levels of this: the quality of the relationship between the federal government and the provinces, and then there's the relationship between the federal government and the people of Canada and whether they're really demanding that the agenda that the government was elected on is actually delivered. If they are, the provincial governments will know that and that makes it more likely the negotiation will succeed.

Thank you very much, Sir Michael!

Thank you!



Information for Government of Canada employees: Coronavirus disease (COVID-19)

Updates for employees on COVID-19, telework, advisories, messages and resources to share.
COVID-19: Learning resources

COVID-19: Learning resourcesNew

Learning resources to support public servants during the COVID‑19 pandemic.

This website is continually being updated in response to your feedback.

Let us know what you think of it.
Which platform is your comment about:
Enter your email if you would like a reply:
Date modified: