Many thanks for the thoughtful and constructive response. I agree with many of your comments.
(First, I note that our published text does not include the sentence you quoted:
We think the base rate is that most efforts to improve governance have ended in failure or worse
Instead it says this:
Ultimately, we think the base rate is that many efforts to improve policy in the last 50 years have ended in failure or worse.
We agreed with your comments on that and amended accordingly before publishing. Thank you again for giving them.)
Responding to your two main points in reverse order:
This draft of the bill makes much more sense when you see it as a campaigning tool, a showcase of ideas. This is a private members bill (PMB). PMBs are primarily campaigning techniques to build support and spark debate. It will not be passed through parliament (in its current form).
1. We did not intend to criticize the Bill as a campaigning device. We intended to express doubts that it should be enacted, and it seems there is more consensus on that than we thought. We expressly said that we never meant to criticize anyone’s actions. And we agree with you about the minimal likelihood of this particular Bill being enacted, although that does not prevent it being the basis of a subsequent bill, so it still matters what the Bill contains and whether those provisions should become law.
2. However, even as a campaigning device, it would be helpful to know what elements of the Bill have won support. Has it built more support for workable and beneficial measures, or for a range of populist measures that will be harmful? We do not have the data to judge but given the weight of the latter in the draft, I am concerned.
If the counterfactual to this Bill is no Bill, I think my view (with considerably lower confidence than on the enactment question) is that no Bill may on balance have been better, insofar as it may create or reinforce unhelpful Schelling points for damaging ideas. That relates to my doubts about the substantive theory of change:
None of the criticisms in the post really pertain to the core elements of the bill. The theory of change for the bill is: government doesn’t make long term plans > tell government to make long term plans (i.e. set a long term vision and track progress towards it) > then government will make long term plans. This approach has had research and thought put into it.
Again, let me stress that I highly welcome the ends and I am keen to discuss more about how to achieve them. My sole concern is with this particular theory of change.
1. If we are looking at the Bill as legislation then what the ‘core elements of the bill’ are could have different meanings. In this regard we prefer to look at likely outcomes rather than intentions. There are many other elements of the Bill that we think would likely make the Bill net damaging to welfare if enacted as law.
2. I strongly support long term planning by the Government, so long as it is done with nearly enough epistemic humility. It is not true that the Government doesn’t make long term plans. There are many sectors where the Government does make or has made long term plans. At present, they are often extraordinarily badly designed.
a. For example, since the 1930s the Government has planned in one way or another to try to push against economic gravity (agglomeration effects) to drive jobs away from higher productivity areas of the country, at enormous costs to welfare. I wrote a casual summary of some such attempts here.
b. The planning system in general attempts to forecast ‘need’ and various other metrics in at least the medium term, and in the cases of some infrastructure for the long term. Those forecasts are often circular, not least because the population movements will partly depend upon the amount of housing and other infrastructure that gets built.
c. The negative effects caused by the bad design of the current planning system have, I estimate, probably damaged welfare per head by at least 10%, while increasing pollution and inequality, among many other problems, as my co-authors and I wrote about here.
d. The Government makes long term plans for infrastructure, of which HS2 is one of the largest examples. While I strongly support building infrastructure in general, I have profound concerns about whether HS2 itself is the best use of the many scores of billions of pounds that will be spent on it.
e. The Government makes long term plans for defence. But I have seen private estimates that the UK’s defence capabilities if attacked are vastly less strong than is assumed by the Government.
f. In relation to the specific institutions you name in the excerpt you quote, I agree IPA may have helped (although we need to do better: infrastructure planning in the UK in general is still worse than many other European countries). But the BoE has for decades provided banks with a range of implicit subsidies. As a result, the banks have eye-watering degrees of leverage and are still profoundly unstable, risky and rent-laden. The BoE has not ensured any plausible plans to resolve major banks in the event of a financial crisis to avoid the necessity of yet another bailout to prevent the collapse of the financial system.
Insofar as the institutions you mention have been helpful, it is worth noting that they are sector-specific. I think there is substantial risk that any general, non-sector-specific Government long-term plans may end up having effects as disastrous as those of the Barlow Report.
So in general I think the Government as currently made up seems (a) lacking in epistemic humility, (b) not good at picking tractable areas in which to make long term plans, and (c) frequently incapable of competently executing on them.
Therefore I think that getting the Government as currently formed to select and act on long term plans, unless the subjects of those long term plans are very carefully guided, could be highly damaging to the welfare of future generations, at least if we look at the long term plans that it has previously created and executed upon that have had the most economic impact over the last 50 years.
So I think there is considerable risk that wellbeing goals set through public consultation will be unattainable or even outright damaging. Goals set by UK governments often seem to be pure responses to short-term electoral or populist incentives and frequently are not even as rational as that.
Response to specific points
I agree that the analysis done in advance of the Bill has been thoughtful and I welcome it. However I think we need to go much deeper to get any confidence about likely outcomes. Mushtaq Khan might be a good place to start in analysing what would be workable given existing UK institutions, and I think his papers illustrate the depth of analysis that it would be useful to have. I welcome your interest in sector-specific approaches and I think there is scope for an enormous amount of progress here.
Conclusion
I would welcome trials of this approach, despite many historically disastrous attempts at long term planning by the UK Government, but the proposals for long term planning in this Bill seem to be a proposal for implementation rather than a trial. I would prefer a trial to be conducted on some selective basis so that we can measure the counterfactual. We could try selecting different sectors, or we could try imposing such requirements on a random selection of mayors, for example. We might also do more in depth study of the many disastrous attempts at long-term planning that the UK Government has previously undertaken, and attempt to build in legislative safeguards against as many of those cases as possible before seeking implementation. But, as I said above, I think a sectoral approach to getting more long-term perspectives and actions in government may be much more productive.
Again, I think you for all your work on this and for the constructive feedback, and I am pleased that we agree on many goals. I hope we can continue this discussion and would welcome collaboration.
Hi John, Thank you for the healthy debate. I am finding it very interesting.
I just want to pick up on one point that is perhaps an important crux where I think my views are quite different from yours. You say:
I think that getting the Government as currently formed to select and act on long term plans, unless the subjects of those long term plans are very carefully guided, could be highly damaging
You seem to be saying that the bill would encourage government to do more long-term decision making but government decision making about the future is just so terrible that we should not encourage it to happen (at least any time soon, at least in any general way).
Insofar as there is a problem of political short-termism (which I think there is considerable evidence for) this line of reasoning seems to suggest that we should not fix that problem (at least for now, in any general sense). I don’t buy this. I tend to apply a very high standard of evidence to arguments against fixing a problem and/or in favour of making systems worse in some way. (As a thought experiment if the government regularly made long-term plans and had a clear way to do it, would you want to stop that happening?)
Furthermore, your evidence for this claim is not strong – it appears primarily to rest on a list of anecdotal stories of failures in long-term planning stretching back over 50 years. But this appears one-sided. I can rattle of an equally long list of times when I think long-term planning has gone or is going well (in reducing regulation, in financial stability, in cyber security, in energy, in nuclear decommissioning, in forestry, in climate, in flood defence, in sewers, etc, etc). There is certainly no universal rule that government long-term planning has to go badly. I also think we should recognise that the significant development of government capability in some long term planning tools over recent years in government such as the impressive Futures Toolkit or the cross-Whitehall heads of horizon scanning project (I think these also serve as examples of non-sector specific approaches).
This is perhaps an important crux. It seems to me that in addition to your (often valid) criticisms of the drafting and some of the fluff in the bill you also have a belief that the bill should not be doing what the bill sets out to do – to generally make government more long-term! I am not sure how to resolve this disagreement. Like I said, I apply a high standard of evidence to such claims that we should not fix problems. But keen to hear your views.
Many thanks for the thoughtful and constructive response. I agree with many of your comments.
(First, I note that our published text does not include the sentence you quoted:
Instead it says this:
We agreed with your comments on that and amended accordingly before publishing. Thank you again for giving them.)
Responding to your two main points in reverse order:
1. We did not intend to criticize the Bill as a campaigning device. We intended to express doubts that it should be enacted, and it seems there is more consensus on that than we thought. We expressly said that we never meant to criticize anyone’s actions. And we agree with you about the minimal likelihood of this particular Bill being enacted, although that does not prevent it being the basis of a subsequent bill, so it still matters what the Bill contains and whether those provisions should become law.
2. However, even as a campaigning device, it would be helpful to know what elements of the Bill have won support. Has it built more support for workable and beneficial measures, or for a range of populist measures that will be harmful? We do not have the data to judge but given the weight of the latter in the draft, I am concerned.
If the counterfactual to this Bill is no Bill, I think my view (with considerably lower confidence than on the enactment question) is that no Bill may on balance have been better, insofar as it may create or reinforce unhelpful Schelling points for damaging ideas. That relates to my doubts about the substantive theory of change:
Again, let me stress that I highly welcome the ends and I am keen to discuss more about how to achieve them. My sole concern is with this particular theory of change.
1. If we are looking at the Bill as legislation then what the ‘core elements of the bill’ are could have different meanings. In this regard we prefer to look at likely outcomes rather than intentions. There are many other elements of the Bill that we think would likely make the Bill net damaging to welfare if enacted as law.
2. I strongly support long term planning by the Government, so long as it is done with nearly enough epistemic humility. It is not true that the Government doesn’t make long term plans. There are many sectors where the Government does make or has made long term plans. At present, they are often extraordinarily badly designed.
a. For example, since the 1930s the Government has planned in one way or another to try to push against economic gravity (agglomeration effects) to drive jobs away from higher productivity areas of the country, at enormous costs to welfare. I wrote a casual summary of some such attempts here.
b. The planning system in general attempts to forecast ‘need’ and various other metrics in at least the medium term, and in the cases of some infrastructure for the long term. Those forecasts are often circular, not least because the population movements will partly depend upon the amount of housing and other infrastructure that gets built.
c. The negative effects caused by the bad design of the current planning system have, I estimate, probably damaged welfare per head by at least 10%, while increasing pollution and inequality, among many other problems, as my co-authors and I wrote about here.
d. The Government makes long term plans for infrastructure, of which HS2 is one of the largest examples. While I strongly support building infrastructure in general, I have profound concerns about whether HS2 itself is the best use of the many scores of billions of pounds that will be spent on it.
e. The Government makes long term plans for defence. But I have seen private estimates that the UK’s defence capabilities if attacked are vastly less strong than is assumed by the Government.
f. In relation to the specific institutions you name in the excerpt you quote, I agree IPA may have helped (although we need to do better: infrastructure planning in the UK in general is still worse than many other European countries). But the BoE has for decades provided banks with a range of implicit subsidies. As a result, the banks have eye-watering degrees of leverage and are still profoundly unstable, risky and rent-laden. The BoE has not ensured any plausible plans to resolve major banks in the event of a financial crisis to avoid the necessity of yet another bailout to prevent the collapse of the financial system.
Insofar as the institutions you mention have been helpful, it is worth noting that they are sector-specific. I think there is substantial risk that any general, non-sector-specific Government long-term plans may end up having effects as disastrous as those of the Barlow Report.
So in general I think the Government as currently made up seems (a) lacking in epistemic humility, (b) not good at picking tractable areas in which to make long term plans, and (c) frequently incapable of competently executing on them.
Therefore I think that getting the Government as currently formed to select and act on long term plans, unless the subjects of those long term plans are very carefully guided, could be highly damaging to the welfare of future generations, at least if we look at the long term plans that it has previously created and executed upon that have had the most economic impact over the last 50 years.
So I think there is considerable risk that wellbeing goals set through public consultation will be unattainable or even outright damaging. Goals set by UK governments often seem to be pure responses to short-term electoral or populist incentives and frequently are not even as rational as that.
Response to specific points
I agree that the analysis done in advance of the Bill has been thoughtful and I welcome it. However I think we need to go much deeper to get any confidence about likely outcomes. Mushtaq Khan might be a good place to start in analysing what would be workable given existing UK institutions, and I think his papers illustrate the depth of analysis that it would be useful to have. I welcome your interest in sector-specific approaches and I think there is scope for an enormous amount of progress here.
Conclusion
I would welcome trials of this approach, despite many historically disastrous attempts at long term planning by the UK Government, but the proposals for long term planning in this Bill seem to be a proposal for implementation rather than a trial. I would prefer a trial to be conducted on some selective basis so that we can measure the counterfactual. We could try selecting different sectors, or we could try imposing such requirements on a random selection of mayors, for example. We might also do more in depth study of the many disastrous attempts at long-term planning that the UK Government has previously undertaken, and attempt to build in legislative safeguards against as many of those cases as possible before seeking implementation. But, as I said above, I think a sectoral approach to getting more long-term perspectives and actions in government may be much more productive.
Again, I think you for all your work on this and for the constructive feedback, and I am pleased that we agree on many goals. I hope we can continue this discussion and would welcome collaboration.
Hi John, Thank you for the healthy debate. I am finding it very interesting.
I just want to pick up on one point that is perhaps an important crux where I think my views are quite different from yours. You say:
You seem to be saying that the bill would encourage government to do more long-term decision making but government decision making about the future is just so terrible that we should not encourage it to happen (at least any time soon, at least in any general way).
Insofar as there is a problem of political short-termism (which I think there is considerable evidence for) this line of reasoning seems to suggest that we should not fix that problem (at least for now, in any general sense). I don’t buy this. I tend to apply a very high standard of evidence to arguments against fixing a problem and/or in favour of making systems worse in some way. (As a thought experiment if the government regularly made long-term plans and had a clear way to do it, would you want to stop that happening?)
Furthermore, your evidence for this claim is not strong – it appears primarily to rest on a list of anecdotal stories of failures in long-term planning stretching back over 50 years. But this appears one-sided. I can rattle of an equally long list of times when I think long-term planning has gone or is going well (in reducing regulation, in financial stability, in cyber security, in energy, in nuclear decommissioning, in forestry, in climate, in flood defence, in sewers, etc, etc). There is certainly no universal rule that government long-term planning has to go badly. I also think we should recognise that the significant development of government capability in some long term planning tools over recent years in government such as the impressive Futures Toolkit or the cross-Whitehall heads of horizon scanning project (I think these also serve as examples of non-sector specific approaches).
This is perhaps an important crux. It seems to me that in addition to your (often valid) criticisms of the drafting and some of the fluff in the bill you also have a belief that the bill should not be doing what the bill sets out to do – to generally make government more long-term! I am not sure how to resolve this disagreement. Like I said, I apply a high standard of evidence to such claims that we should not fix problems. But keen to hear your views.