It seems to me that the entire running theme of this post greatly underestimates the uncertainty of any feasible answer to this question, and that it also greatly overplays the strength of this relationship.
In fairness to Carl, he said that he’ll be addressing the question of what the relationship between purchased votes and policy outcomes in a future post.
In modern Western political systems, the effective role of elected politicians is far smaller and subject to much stronger and more complicated constraints than people who care about day-to-day politics commonly imagine.
What evidence do you have to support this claim?
Another source of great uncertainty is the relation between policy outcomes and the kinds of consequences by which you measure their desirability. (To the extent that these measures are even defined clearly—even if all the facts are known, in many situations there is no obviously correct definition of “saving a life,” and different definitions will lead to very different evaluations.)
Agreed, but despite the uncertainty, the magnitude of the potential impact may be sufficient to justify focus on policy outcomes. See Nick Beckstead’s comment #15 at the Singularity Summit posting on the GiveWell blog.
To answer your question fully, I would have to expound my entire theory of the modern state, which would unfortunately require much more time and space than can be dedicated to a blog comment. So what I write will be very cursory, simplified, and incomplete.
The basic insight is that elected politicians are transitory and in constant danger of having their careers destroyed by bad PR, while the bureaucrats are entrenched like the rock of Gibraltar, constantly running circles around politicians and preventing them from doing anything that deviates significantly from the direction in which things are carried by the bureaucratic inertia. Politicians lack any means to dislodge the bureaucrats, who can in turn make their life miserable in many different ways. In case there’s a direct conflict, the politician loses without exception. The only sensible strategy, which successful politicians inevitably follow, is to simply give up any thought of such conflict.
Of course, the bureaucrats won’t mind if politicians do things that create more bureaucracy, but even in that case, the actual consequences of such measures are principally in the hands of bureaucrats, not politicians. Legislation is nowadays typically written in a long-winded and extremely vague style, leaving it up to the bureaucracy and to some extent the judiciary to shape it into actual policy.
This simple view omits the crucial roles played by the judiciary and by various other centers of non-elective power whose social, organizational, and financial structure effectively blends into the government even though they’re theoretically not part of it, such as the mainstream media, academia, and various for-profit and non-profit nominally private institutions. However, these merely present additional limitations to the power and influence of elected politicians, and they typically operate in concert with the bureaucrats, giving them some of their crucial leverage against politicians. It also ignores some (mostly vestigial) limited ways in which politicians can sometimes exert direct control over things.
On the whole, this is an immensely complex and controversial topic. However, any plan for influencing things by electing politicians must recognize this state of affairs, or otherwise it completely loses touch with reality.
To answer your question fully, I would have to expound my entire theory of the modern state, which would unfortunately require much more time and space than can be dedicated to a blog comment.
If you ever do write this up, I’d be very interested to read it. Incidentally, your theory sounds quite similar to Yes Minister, incidentally also the source of one of my favourite quotes about politics:
“Politician’s logic: We must do something. This is something. Therefore we must do it.”
Incidentally, your theory sounds quite similar to Yes Minister.
Oh, yes! In full seriousness, while the plot and dialogue in that series are exaggerated for comic effect, it depicts the structure of modern governments with more essential accuracy than probably any academic work of political science.
There’s a good reason for it. The series was vaguely based on the published diaries of Richard Crossman, who might be the only modern-age politician who published an unadorned day-to-day diary of his work instead of a customary auto-hagiography of the sort we usually get from politicians.
Here’s the opening passage of his diaries (with the source of the legendary “Yes, Minister” title highlighted by me):
I was appointed Minister of Housing on Saturday, October 17th, 1964. Now it is only the 22nd… and already I realize the tremendous effort it requires not to be taken over by the Civil Service. My Minister’s room is like a padded cell, and in certain ways I am like a person who is suddenly certified a lunatic and put safely into this great, vast room, cut off from real life and surrounded by male and female trained nurses and attendants. When I am in a good mood they occasionally allow an ordinary human being to come and visit me; but they make sure that I behave right, and that the other person behaves right; and they know how to handle me. Of course, they don’t behave quite like nurses because the Civil Service is profoundly deferential - ‘Yes, Minister! No, Minister! If you wish it, Minister!’ and combined with this there is a constant preoccupation to ensure that the Minister does what is correct. The Private Secretary’s job is to make sure that when the Minister comes into Whitehall he doesn’t let the side or himself down and behaves in accordance with the requirements of the institution. It’s also profoundly true that one has only to do absolutely nothing whatsoever in order to be floated forward on the stream...
How independent is your theory from that of Mencius Moldbug?
My impression is that the theory could be said to be more correct if it was clear what variables are treated as dependent and as independent, or as agentic and detemined. As is, the theory is underdetermined.
How independent is your theory from that of Mencius Moldbug?
I’ve read lots of stuff written by Moldbug, and his ideas have influenced me somewhat, though the exact scope of this influence is hard for me to disentangle. On the other hand, the basic point from the above comment, i.e. that electoral politics is largely irrelevant compared to the entrenched bureaucracy and various nominally non-state entities on its periphery, was my opinion long before I first heard of him. Then again, by that point I had been independently exposed to, and influenced by, a lot of the literature that he draws on.
In any case, I’ve never observed anything that would contradict significantly even the rough picture I paint in my above comment.
It looks to me like the Carter and Reagan administrations in the US and the Thatcher administration in the UK really did greatly cut back the bureaucracy fairly successfully. Also, like the bureaucrats don’t decide what wars get fought, etc.
It looks to me like the Carter and Reagan administrations in the US and the Thatcher administration in the UK really did greatly cut back the bureaucracy fairly successfully.
Carter’s (of all people!) deregulation of airlines and the subsequent phase-out of the CAB was indeed a rare example of politicians effectively shutting down an entire bureaucratic agency. The rest of Carter’s record is very different, though; for one, his administration created the federal departments of education and energy. (And frankly, I’d be surprised if any major CAB bureaucrats actually got laid off rather than transferred to equally cushy positions.)
Regarding Reagan, I disagree. His ascent was indeed seen back then—with hope or horror, depending on whom you asked—as a reactionary tsunami that would sweep away huge parts of the federal bureaucracy, and he openly campaigned on this sentiment. Yet, in practice, he achieved almost nothing. This tremendous populist momentum crashed against the Washington bureaucracy while barely making a dent in it. Reagan didn’t even manage to eliminate the fledgling education and energy departments that Carter had just created, which he promised explicitly; he also failed in his later plan to merely trim them somewhat and turn them into subdivisions of other departments.
With Thatcher, it’s a similar story. Public spending went up during most of her years in office. The reductions in civil service size under her might look significant—until you realize that most of it was due to eliminating blue-collar jobs from the civil service payrolls and replacing them with subcontracting (see Figures 3 and 4 in this paper). (Not to mention how well bureaucrats know to cook the numbers to make them look like “reductions” where there’s no such thing really going on. It’s classic “Yes Minister” business.)
Worse yet, when politicians manage to score some partial victory in reducing the size and scope of bureaucracy, it typically turns out to be a random lucky victory that throws a wrench into some part of the government machinery, rather than a sensible reorganization. This part then usually malfunctions with visible bad consequences—resulting in rushed measures to fix the situation by employing even more bureaucracy than there had been in the first place. (And even more bad PR for “deregulation.”)
Also, like the bureaucrats don’t decide what wars get fought, etc.
Yes, war, and to some limited extent foreign policy, are among the exceptions I mentioned, where elected politicians in the U.S. can still influence things significantly. But while a president and a willing Congress can start a war without much say from the bureaucrats, the bureaucrats, with the help from their allies in the media, judiciary, etc., still have the power to make the success in this war impossible, and to turn it into a slow-motion career disaster for the politicians involved.
There is also the constraints from international agreements (ACTA etc).
And just trying to manage to cope in the world economy, you can’t make your country less friendly to big businesses than the states similar to you, otherwise they will choose to move their research facilities/factories to the other places.
In fairness to Carl, he said that he’ll be addressing the question of what the relationship between purchased votes and policy outcomes in a future post.
What evidence do you have to support this claim?
Agreed, but despite the uncertainty, the magnitude of the potential impact may be sufficient to justify focus on policy outcomes. See Nick Beckstead’s comment #15 at the Singularity Summit posting on the GiveWell blog.
multifoliaterose:
To answer your question fully, I would have to expound my entire theory of the modern state, which would unfortunately require much more time and space than can be dedicated to a blog comment. So what I write will be very cursory, simplified, and incomplete.
The basic insight is that elected politicians are transitory and in constant danger of having their careers destroyed by bad PR, while the bureaucrats are entrenched like the rock of Gibraltar, constantly running circles around politicians and preventing them from doing anything that deviates significantly from the direction in which things are carried by the bureaucratic inertia. Politicians lack any means to dislodge the bureaucrats, who can in turn make their life miserable in many different ways. In case there’s a direct conflict, the politician loses without exception. The only sensible strategy, which successful politicians inevitably follow, is to simply give up any thought of such conflict.
Of course, the bureaucrats won’t mind if politicians do things that create more bureaucracy, but even in that case, the actual consequences of such measures are principally in the hands of bureaucrats, not politicians. Legislation is nowadays typically written in a long-winded and extremely vague style, leaving it up to the bureaucracy and to some extent the judiciary to shape it into actual policy.
This simple view omits the crucial roles played by the judiciary and by various other centers of non-elective power whose social, organizational, and financial structure effectively blends into the government even though they’re theoretically not part of it, such as the mainstream media, academia, and various for-profit and non-profit nominally private institutions. However, these merely present additional limitations to the power and influence of elected politicians, and they typically operate in concert with the bureaucrats, giving them some of their crucial leverage against politicians. It also ignores some (mostly vestigial) limited ways in which politicians can sometimes exert direct control over things.
On the whole, this is an immensely complex and controversial topic. However, any plan for influencing things by electing politicians must recognize this state of affairs, or otherwise it completely loses touch with reality.
If you ever do write this up, I’d be very interested to read it. Incidentally, your theory sounds quite similar to Yes Minister, incidentally also the source of one of my favourite quotes about politics:
mattnewport:
Oh, yes! In full seriousness, while the plot and dialogue in that series are exaggerated for comic effect, it depicts the structure of modern governments with more essential accuracy than probably any academic work of political science.
There’s a good reason for it. The series was vaguely based on the published diaries of Richard Crossman, who might be the only modern-age politician who published an unadorned day-to-day diary of his work instead of a customary auto-hagiography of the sort we usually get from politicians.
Here’s the opening passage of his diaries (with the source of the legendary “Yes, Minister” title highlighted by me):
Wow, so it is accurate for the same reason as the The Wire (based on a study of reality), that’s awesome.
How independent is your theory from that of Mencius Moldbug? My impression is that the theory could be said to be more correct if it was clear what variables are treated as dependent and as independent, or as agentic and detemined. As is, the theory is underdetermined.
MichaelVassar:
I’ve read lots of stuff written by Moldbug, and his ideas have influenced me somewhat, though the exact scope of this influence is hard for me to disentangle. On the other hand, the basic point from the above comment, i.e. that electoral politics is largely irrelevant compared to the entrenched bureaucracy and various nominally non-state entities on its periphery, was my opinion long before I first heard of him. Then again, by that point I had been independently exposed to, and influenced by, a lot of the literature that he draws on.
In any case, I’ve never observed anything that would contradict significantly even the rough picture I paint in my above comment.
It looks to me like the Carter and Reagan administrations in the US and the Thatcher administration in the UK really did greatly cut back the bureaucracy fairly successfully. Also, like the bureaucrats don’t decide what wars get fought, etc.
MichaelVassar:
Carter’s (of all people!) deregulation of airlines and the subsequent phase-out of the CAB was indeed a rare example of politicians effectively shutting down an entire bureaucratic agency. The rest of Carter’s record is very different, though; for one, his administration created the federal departments of education and energy. (And frankly, I’d be surprised if any major CAB bureaucrats actually got laid off rather than transferred to equally cushy positions.)
Regarding Reagan, I disagree. His ascent was indeed seen back then—with hope or horror, depending on whom you asked—as a reactionary tsunami that would sweep away huge parts of the federal bureaucracy, and he openly campaigned on this sentiment. Yet, in practice, he achieved almost nothing. This tremendous populist momentum crashed against the Washington bureaucracy while barely making a dent in it. Reagan didn’t even manage to eliminate the fledgling education and energy departments that Carter had just created, which he promised explicitly; he also failed in his later plan to merely trim them somewhat and turn them into subdivisions of other departments.
With Thatcher, it’s a similar story. Public spending went up during most of her years in office. The reductions in civil service size under her might look significant—until you realize that most of it was due to eliminating blue-collar jobs from the civil service payrolls and replacing them with subcontracting (see Figures 3 and 4 in this paper). (Not to mention how well bureaucrats know to cook the numbers to make them look like “reductions” where there’s no such thing really going on. It’s classic “Yes Minister” business.)
Worse yet, when politicians manage to score some partial victory in reducing the size and scope of bureaucracy, it typically turns out to be a random lucky victory that throws a wrench into some part of the government machinery, rather than a sensible reorganization. This part then usually malfunctions with visible bad consequences—resulting in rushed measures to fix the situation by employing even more bureaucracy than there had been in the first place. (And even more bad PR for “deregulation.”)
Yes, war, and to some limited extent foreign policy, are among the exceptions I mentioned, where elected politicians in the U.S. can still influence things significantly. But while a president and a willing Congress can start a war without much say from the bureaucrats, the bureaucrats, with the help from their allies in the media, judiciary, etc., still have the power to make the success in this war impossible, and to turn it into a slow-motion career disaster for the politicians involved.
This is my worldview as well.
A couple of other constraints:
There is also the constraints from international agreements (ACTA etc).
And just trying to manage to cope in the world economy, you can’t make your country less friendly to big businesses than the states similar to you, otherwise they will choose to move their research facilities/factories to the other places.