Re-release: Power Transitions

Speaker A: 00:00:00

Hello and welcome to the Cognitive Engineering Podcast produced by me, Brace McGruer for Aleph Insights. In this series of podcasts we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Chris Wragg and Nick Hare of Aleph Insights and this week we're discussing the potential problems Joe Biden might have when he comes to power. Chris, lead us in.

Speaker B: 00:00:29

Obviously you have a new incoming president, the president-elect, and they are in the process, or certainly some of the process has been hampered by dispute around the result of the election and the power transition. But essentially it got me thinking about this idea of when one regime comes in and takes over from another about the potential issues you have occurring within a system when control is handed over from one set of people to another set of people and whether there are recipes for success or whether the context in which it happens means that sometimes power transitions are doomed to be inefficient

Speaker A: 00:01:27

and problematic. Okay, right, okay, that's teed it all up quite nicely. Where do we want to start?

Speaker B: 00:01:33

Yeah, well I think the main thing that is different probably about this presidential transition that is not the case with previous ones. I mean there have been some that have been ill-tempered and where the outgoing president has been obstructive or not turned up to the inauguration of the next president. But this one is, I wouldn't say it's unique, but it's definitely

Speaker A: 00:02:01

in a small sample. So we're going to start off by talking about then this particular transition. I think famously, was it Clinton's people stole all the W's from the keyboards when George W Bush was coming in? Sorry, carry on Chris, I interrupted you.

Speaker B: 00:02:22

No, no, that's, yeah, that's, I mean at least that's hijinks rather than sort of, you know, actually actively trying to disrupt all the policies of the incoming government. But you always get this issue when, you know, in the US system, you know, unless, you know, well even when it's from when the incoming president is from the same party to, you know, somebody who's just finished their second term or something, there is still a change of direction. But I think what is particularly interesting about this one is that it has been so acrimonious and the outgoing president has been the least cooperative, I think, that, you know, that you can sort of think of in modern history and that the incoming president will want to change lots of things quite dramatically. So you've got somebody sort of digging in and causing problems and you've got somebody who's wanting to change things, you know, quite dramatically. And so I think that's an interesting dynamic and I think there are lots of, you know, things we can sort of look at historically to kind of give some insight into power transition.

Speaker A: 00:03:41

Yeah. And also I think it's quite interesting in a country like the States, like America, where they've got this power transition every four years, and at least the last couple of hundred years or so, there's these rules in place. And yet still there are some issues, it seems, if you've got one sort of particularly sort of, what's the best way to put it? There you go. Thank you. So even in a system that is a system, it's got some safeguards in place.

Speaker C: 00:04:10

That's interesting. Nick? Yeah, I think, well, one of the things I was interested in is whether this perception is correct, that, you know, sort of transitions are dangerous or, you know, uncertain or hard in some way. It's hard to tell in the US, right? Because as you say, they have a four year cycle. And that means that, you know, you can't look at the length, for example, of a president's tenure to try and make inferences about whether or not, you know, things are more likely to go wrong at the beginning or whatever. Although, obviously, you do have this idea of the first hundred days being a kind of key time. But that's really, I think that's more about kind of being able to use your political capital rather than your tenure being more uncertain at the beginning of your time in office. I did actually have a quick look at US crime rates. I thought, well, maybe somehow, you know, if there's a sense of there being more instability around the beginning or the end of a, you know, presidential term, we might be able to see that picked up in crime rates. In fact, there's really no electoral cycle there at all. But what I did have a look at was some of the lengths of the durations in countries where there isn't that same electoral cycle. So, first of all, I looked at the UK. Now, what you can do is sort of plot, well, look at what's called a hazard function where you look at a sort of failure rate over time. You know, you look at, you take a large population, you see how many of that population, you know, period one, suffer a certain thing, and then in period two, and then in period three. Now, if beginnings, ends doesn't really matter, what was it sort of if whereabouts you are on in your tenure doesn't affect the probability of failure, then what you would get is an exponential function. You'd have the same sort of decay rate over time. So the number of prime ministers who'd been in power for nine years, but not for 10 years would be the same proportion as you know, those who'd been in power for two years, but not for three years. So I looked at the UK and there is, you know, it's definitely the case that there is sort of generally a higher failure rate for the early years. So 25% of prime ministers don't make it out of year one. So I'm talking about since Walpole here. So 25% don't make it out of year one. 21% if you've been in year one, make it to year two. But by the time you get to sort of five years, there's been 24 prime ministers who've been in power for five years. Only 11% of them fail to make it to six years. And you know, once you get to sort of eight, nine years, it's 11, 13%. So there is some evidence that the decay rate falls off over time. So what that suggests is, yeah, that actually, you know, the beginning your first period, your first few years in office are really kind of crucial. And the time we're actually most likely to be kicked out again. And I also thought just for interest, another country famously, which, you know, famously has had a lot of prime ministers, which is Italy. They've had hundreds, I mean, you know, and the failure rate for the first four years, right, for if you look at the average failure rate over the first four years of an Italian prime minister's tenure in office, there's a failure rate of about 5.8% per month. But over the next two years, it falls to 2.2%. So again, same thing, you know, if you're an Italian prime minister, you've got a higher probability of staying in office. Admittedly, only about only four prime ministers have ever done that, one of them being Mr. Benito Mussolini. Anyway, sorry, Frazee, you're trying to say something. Yeah, well, I know Chris wants to

Speaker A: 00:08:16

come in, but I'd be interested to see how one this relates to transition. Okay. But secondly, it'd be interesting to see if, you know, short-lived prime ministers, how soon or not that is after a long-lived prime minister. So you might have long, short, short, short, long,

Speaker C: 00:08:32

short, short, short, something like that. I don't know. Anyway, Chris. Sorry, but do you want me to make it explicit why that's related to the hypothesis? It's just, okay, so it's just the idea that, you know, if power transitions, you are just taking over, if that's a particularly tricky or difficult time, you would expect to see a lot of people fail in the first year or two, say, compared to the failure rate of prime ministers beyond that. And what I'm saying is that that is backed up. So now it doesn't mean it's the only hypothesis consistent with that evidence, but it is, you know, it is the case. There isn't the same failure rate over time. Basically, you know, if you can hang on, if you can make it through that first bit, you're safer, which, you know, that is merely a way of saying that the early part of your tenure is the dangerous part.

Speaker B: 00:09:23

Transition. I think that's, you know, an interesting hypothesis, because obviously, what you then do when you're trying to set up a system to, I suppose, try and minimize that instability. So like the US presidency has fixed terms, right? So you can't, you know, other than by impeachment or death, or, you know, a couple of other processes, you can't unelect a president before the end of their term. Whereas the UK system, now we've got the Fixed Parliament Act, but even within that, we've still had somebody call a general election, in fact, two, you know, outside of the general, you know, five-year fixed terms that were put in place. So I think that's an interesting way of looking, you know, at the fact that you do get greater instability at the beginning of kind of, you know, somebody ascending to power. And so if you're trying to alleviate that, obviously, you would set up a system where you do have these fixed terms. But I suspect that brings other problems. And if you look at other kinds of systems, so we've talked about democratic systems. But if you look at what was the majority political system for a long time, if you look at, you know, hereditary monarchies and transition across those, you've got all sorts of problems, which probably have parallels in democratic systems, but are nowhere near as bad. So if you look at the monarchs of England or Great Britain, from William I through, so from William the Conqueror through to now, 40 periods of monarchy. And if you look at how many of those are contested, I think it's fair to say William the Conqueror's ascendancy to the throne was definitely contested. But I went through and I had a look at these. And it's quite astonishing because you think the English monarchy is pretty stable, generally, as a system. It's obviously existed

Speaker C: 00:11:53

for a very long time. But you also think, well, aren't there all these kind of convoluted rules

Speaker B: 00:11:57

about, you know, agnostic? And obviously, each time there was an issue, they then tried to bring in some, you know, some amendment to try and resolve that in some way. But of the transitions between those monarchs, I went through and sort of made a judgment on them. And 35% of them were violently contested in some way. So, you know, there was either an uprising at the time, somebody deposed the king, there was a, you know, it was the result of a civil war, and so on and so forth. And so, you know, that suggests that that kind of system where the only way out of it is the death of the reigning monarch, or some form of, you know, deposing them is, you know, so you've effectively got their lifetime, really, as the tenure and the transition happens at the end of that lifetime, is clearly not a stable way to transition things. What I didn't do, which would be very interesting to do, I think, is to look at that notion of whether or not you get, you know, rapidity, you know, when you get a long, because intuitively, you think a long reign creates stability, right? You know, and there are a couple of instances where kings swap over, where actually, you know, the king is deposed momentarily, and then they're sort of reinstalled, and so on. So, yeah. Okay. Yeah, I'm slightly lost. Where are we? I think

Speaker A: 00:13:44

we've got to the point where, yeah, democracy is definitely better than something like a monarchy.

Speaker B: 00:13:50

Yeah, yeah. I mean, yeah, I mean, just in terms of transition, right? Yeah, well, well, so, so the other thing I looked at first, you know, it because it's a case study in instability, was the Weimar Republic. And the fact that they- I thought you were going to pick on poor Italy again. But anyway, no, no, no, even more tumultuous than post-war Italy was the Weimar Republic. Yeah, where they had 14 chancellors in 14 years, with, yeah, obviously, von Schleicher being replaced by some chap nobody's heard of since. An Austrian guy. Yeah, an Austrian guy. Yeah. So, so that, you know, clearly there, you've got really rapid turnover of power. So there's something about, you know, the rate of turnover, which clearly in itself creates instability in systems like that. I think, I mean, I know that in early Rome, I think tributes served for, for a year, and there were there were two, you know, two of them. And, you know, they sort of shared, shared power. But, but I think it's probably rare to have systems where you have rapid turnover of power, and you have stability. Yeah, I think, and I think it,

Speaker C: 00:15:17

you know, it's interesting thinking about that. I mean, obviously, one of the reasons that the Roman Republic had a system like that, where you where you turn, you deliberately got rid of people didn't matter how good they were, they had to go and be replaced by someone was precisely because they were scared of, you know, the what we've been looking at as an indicator of stability, which is having the, you know, a kind of same person ruling for a long time, gathering a lot of personal power, and, you know, and overshadowing the the sort of legislative power. And, and I, but I think, and I, obviously, the US have a have a similar thing, you know, it's this fear of the fear of it sort of you like kind of too much stability. And the fact that that drags in the all of the problems to do with having a cult of personality, that, you know, it's the flip side

Speaker A: 00:16:07

of having a long and stable reign. I mean, fixed term limits is very common around the world, I think in most countries is probably the case. Yeah, very, very common. Yeah.

Speaker C: 00:16:19

So I've interrupted your flow there, Nick, sorry. No, I was but I was just gonna say, I mean, the on one hand, you can point to procedural reasons why you're gonna get instability, the most obvious one, take Biden, you know, he's gonna have to replace a whole load of officials, obviously, in the US, they have a bit slightly more politicized civil service than we do. And so that means, you know, replacing what we, you know, would be effect, it's a bit like putting ministers in place, you know, in government departments here, but they're sort of more less, they're not really ministers, they're more like senior civil servants who are kind of politically appointed. But so a lot of those, I mean, all of them are going to go and be replaced by new ones. And, but also Biden is not going to really know which of those people he can trust, you know, who is going to be actually carrying out his what he really wants, who's going to be pursuing their own career, who's going to be conniving behind his back, he doesn't know how much he can trust Kamala Harris, or how good she's going to be, there's all kinds of reasons why, you know, you would imagine that when someone gets into power that you things, things could go, you know, a number of different ways. And it takes some time for someone to be able to generate stability in that period. But another alternative hypothesis, which I just think it's worth pointing out is that it's not that start, the beginning of a new administration is particularly risky, necessarily, but what it is, is highly informative, right? So this would be a different hypothesis that reverses the cause and effect, and merely would sort of say, well, the point is that you don't really know how good a leader is going to be when they first get into power. And the reason that you see a lot, you know, a relatively large number of short lived leaders, when you look at, you know, systems where they can choose the length of the leadership, is that you've got to filter out the rubbish ones. And the only way you can do that is by giving them a shot at it. So it's not that, you know, you become more stable as you go on, but it's merely that the people who are rubbish get filtered out, which is another hypothesis, you know, it's not the transition that causes instability is the sort of almost the other way around. Yeah. And I think that again,

Speaker B: 00:18:26

here, history is quite instructive, and other systems are quite instructive, because I feel like I think I've mentioned previously, Mehmed III, the Ottoman Sultan, who upon ascending to the Sultanate, had his 19 brothers murdered. And, you know, that is essentially the executive change that occurs in the US system, just without the throttling, you know. And so you do have this sense of, right, okay, we've got to get rid of people and bring in people who are my, who I patronize, you know, as the leader and who owe their loyalty to me. I think the other thing that's interesting is if you look at systems where you have had ideological flip flopping, right, as you do get in the US system, you know, Obamacare turns up, Trump comes along, cancels Obamacare, and so on. Now, if you think about, yeah, during the sort of English monarchy and the Reformation, and people burning people at the stake on either sides, either as Protestants or Catholics for heresy, you've got, you know, again, another factor there. So you've got this idea of having to, you know, really purge people within your system who follow one ideology, because otherwise, you can't, you know, you can't pursue your sort of policy. And then I think the other factor you've got is internal disloyalty, which Nick was sort of, you know, hinting at, I suppose, a bit, which is, you know, often in monarchies, you get, you know, the heir becoming, you know, sort of agitating because their father won't die. And so, you know, they're kind of kicking their heels. And goodness, we don't have that kind of situation. Yeah, that's right. And, and so, yeah, you've got a number of factors there, which are basically, you know, the fact that when you change over, you've got to sort of change the system to be loyal to you. But you've also got this issue of if you've been around for too long, you know, those beneath you want to take the reins of reins of power. So that probably puts a natural limit on

Speaker C: 00:20:55

things. Yeah. So it's probably one other thing worth mentioning, which is just alluding to, or kind of think of jumping off something you said earlier, Fraser, which was, you know, are we going to see a long followed by some short reins, and then another long one, there is going to be a kind of element of reversion to the mean, right? So which means that you're more, most regimes are bad, right? I mean, on average, you look at all governments over the history of mankind, there's, there are very few, you know, Queen Victoria's or Alexander, Alexander's the Great. And it's, it is just overwhelmingly likely that you're going to get a, you know, a worse person, if you've had a really, really good person in charge, you know, who's managed to hold on to power for a long time, or has been a good leader. I mean, Alexander the Great is a really good example, because, you know, he built this enormous empire bigger than you anything the region had seen. And then as soon as he, you know, the state fell apart as soon as he died, you know, it was more or less just completely fell into, fell apart into different, totally different, you know, the whole country was so bound up with him, it was, you're not going to get another Alexander the Great to turn up, you'd be very lucky to, you know, to have, to have one,

Speaker A: 00:22:09

let alone two in a row. It makes me think of football and football managers in Liverpool, who had what was it? Exactly Shankly, then followed by, was it Paisley? I can't remember. And then on to, and that was really, exactly Dalgleish. And then Jonas messed everything up.

Speaker C: 00:22:28

I think Alexander the Great and Bill Shankly are very similar figures in a way in terms of their impact on world history. We need to finish there. But before we do, a little question. Why do we make predictions for when the Queen dies?

Speaker A: 00:22:45

Well, I've got a different, and usually I'm going to ask you for a prediction. Is Donald Trump going to be at the inauguration in the States?

Speaker B: 00:22:55

he will try and run again in: 2024

Speaker A: 00:23:27

to announce his candidacy for: 2024

Speaker C: 00:23:37

is coming from. But I also think it's hard to tell how much is, you know, Donald Trump playing the game here. I mean, I think with someone like Donald Trump, his personality and the way that, the kind of things that he does for one reason and the things that he does because he wants to are so tightly bound up, it's hard to separate them. But I am thinking about the, you know, the base rate here, which is that it's just so rare for a president not to turn up to the next inauguration. Now, admittedly, Trump has done, you know, has done plenty of weird- Has done the rare thing a few times. Yeah. But I'm going to go with 75% because I just think it seems it would be extremely weird. I mean, on the other hand, maybe it'd be a way of overshadowing events, you know? What's the base rate for Trump taking the weird decision on every turn? Yeah, yeah. It's what happens when that base rate meets the other base rate, you know, which base rate has to shift.

Speaker A: 00:24:38

I think that, I think he will be there, but might do something. I don't know, might do something. I don't think he just would be happy to sit there.

Speaker C: 00:24:47

I think that's the, that's sort of what I'm thinking is that I think he'll want, he won't risk not being in the limelight for a minute. He'll try and steal the show in some way. You know, or at least he might. So I think it's a bit higher than 50% chance he'll be there, I'd say.

Speaker A: 00:25:04

Yeah, yeah. Okay, let's see. All right, we'll stop there. Thank you, as always, for listening to the Cognitive Engineering Podcast. I'm Fraser McGruer, being here with Chris Wragg, Nick Hare of Aleph Insights. Until next time, goodbye.

Speaker B: 00:25:15

Fraser McGruer, did you call yourself that?

Speaker A: 00:25:18

Seriously, I do have trouble with my own name.