The Wonks’ List: the best technical policy reads of 2023

In addition to Grattan Institute’s annual Prime Minister’s Summer Reading List, our Wonks’ List highlights some of 2023’s best technical policy reads, for anyone who wants to take a deeper dive.

In this special podcast, Senior Associate Natasha Bradshaw discusses the Wonks’ List with Associate Elizabeth Baldwin and Health Program Director Peter Breadon.

Read the list

Transcript

​Natasha Bradshaw: In addition to the Grattan Institute’s annual Prime Minister’s Summer Reading List, we also have an additional list for all the nerds out there. Our Wonks’ List highlights some of the year’s best technical policy reads. I’m Natasha Bradshaw and I’m joined by Elizabeth Baldwin and Peter Breadon, all part of our Grattan Book Club.

And today we’re going to talk about our best technical reads of the year.

Liz Baldwin: Last time we were on the podcast together we were talking about Claudia Goldin’s Nobel Prize win and we were having a very nerdy discussion then as well about gender norms and the influence of those on the labor market and now you’re back with part two and you suggested a paper on how gender norms affect men.

Natasha Bradshaw: So as we’ve discussed before, there’s a large stream of literature about gendered norms and their effects on how couples choose to separate paid and unpaid work. And they often have this very strong focus on women’s outcomes in the labor market. But until now, we knew very little about the consequences of masculine norms on men.

So the paper that I want to talk about is called Men, Male Bias, Sex Ratios, and Masculinity Norms, Evidence from Australia’s Colonial Past. And it’s by Victoria Baranoff, Ralph de Haas, and Pauline Grosjean. So what it starts by outlining is that when Britain sent convicts to Australia, it sent many more men than women.

So by the mid 1800s in New South Wales and Tasmania, there were about 28 convict men for every one convict woman. And even after this, most people migrating to Australia were men sort of looking for these economic opportunities. And so the skew towards having many more men than women continued for quite some time in certain parts of the country.

A few years ago, a paper was published by Rose Kata and Pauline Grosjean, aptly titled It’s Raining Men, and they looked at the distribution of these male bias sex ratios across the country and showed how those actually have effects even in the present day. And that’s because gender norms really perpetuate over time.

If you’ve got an area where there’s a lot of men and not many women, the men really have to compete to find partners. And what it means is that women have a lot of power in those relationships. So, way back in the 1800s, Women that lived in these areas that had a lot of men tended to work a lot less outside the home than women in other areas.

And even today, women that live in those areas are still less likely to be in paid work or they work less hours. And that’s because these norms really perpetuate over time. Now, what this new paper looks at is not how these historical, sex ratios affect women, but actually about how they affect men and masculine norms.

And the results are actually just astounding. So, in areas that had more male convicts relative to women, men were more likely to voluntarily enlist for World War I a century later. And even today, these areas have significantly higher rates of violence and sexual assault. Boys, but not girls, are more likely to be bullied at school in these areas.

They have higher rates of male suicide. They also have more avoidance of preventative healthcare by men. So we know from the medical literature that there’s a relationship between sort of adherence to masculine identity and the avoidance of preventative healthcare, so things like screening for prostate cancer, but what they show in the paper is that in these areas with historically male biased sex ratios, there are higher rates of death for men from preventable things like prostate cancer.

They also find that men in these areas are less likely to work in industries that are historically or traditionally female. And so what they show is that over time these norms have just perpetuated through families and through peers so that they’re still harming men even today.

Liz Baldwin: Is this just a sort of historical accident that we have to live with, or are there policies that government can use to kind of address some of those negative consequences you mentioned?

Natasha Bradshaw: Yeah, well it tells us that these historical norms are perpetuated for a very long time and continue to harm men today. So in some ways it serves as a warning that we should avoid these sort of extremely biased sex ratios. So for some countries today that still have practices like sex selective abortions, that skew the sex ratio it does sort of serve as a warning that there are likely to be these quite negative long term effects on men from that.

But even within Australia, it might make us a bit more nervous about, you know, very male biased settings in prisons, in schools or in workplaces that there’s a chance that these do have these quite harmful effects. I think it’s also a lesson when we talk about gender equality and policies that promote you know, women’s workforce participation.

What I think is really nice about this paper is often when we talk about these problems it’s from the perspective that it will help women.

But what we’re seeing here is that actually these effects are also really harmful for men It’s not, you know, a zero sum game, which leads us really nicely into the next paper, Liz, which explores the multi generational effects of economic and family circumstances on beliefs today.

Do you want to tell us a bit about that?

Liz Baldwin: That situation you just described is kind of a classic example of this phenomenon called zero sum thinking, which is the idea that the gains for one group, say, women, must come at the expense of another group, men, in this example. And, you know, therefore that there’s this kind of tension, people are bargaining over a fixed sort of pie or a fixed pot of goods.

And that’s not the case in a lot of policy contexts today. We know that, you know rising productivity, rising economic growth tends to lift all boats. Obviously there are exceptions with inequality, but in very broad terms, a lot of policies that include more people and allow us to use the skills of more people ultimately benefit everyone, but still this kind of thinking persists that a lot of policy puzzles are a zero sum game. And this paper also kind of looks back in history to understand the roots of that sort of thinking today. So it’s set in the United States, and it sent out a big survey to about 20, 000 Americans. And it asked them a series of questions about how they think about certain policy trade offs.

And it also asked them a lot of questions about their family history and where their parents grew up, where their grandparents grew up. And it then sort of mapped people’s beliefs today about that kind of zero sum idea to that family history. And it showed that there are quite strong effects from your family’s circumstances to your beliefs today.

If your family experienced upward mobility across the generations, if your parents did better than your grandparents you’re more likely to think that the rising tide lifts all boats, that you’re less likely to have this kind of zero sum competitive view, because by and large, that wasn’t your family’s experience.

You’ve experienced this sort of positive growth trajectory. And similarly, even if in a broader economic sense, if the area that you grew up in, at the time at which you were growing up, was experiencing a period of rapid economic growth, you’re also less likely to have these beliefs, whereas on the other hand, if your ancestors experienced enslavement or other forms of really deep subjugation, you’re much more likely to have a zero sum, worldview, which is obviously consistent with those experiences.

Natasha Bradshaw: Liz, is this just a story of political alignment? Sort of Democrats versus Republicans, or is there more going on?

Liz Baldwin: It’s slightly correlated with your sort of typical political beliefs, people who have stronger zero sum thinking are more likely to vote democratic, but it’s actually more powerful in explaining the differences within political parties. We know that not every Democrat has the same set of beliefs.

Not every Republican has the same set of beliefs. And what the authors in this paper showed is that if you were a Democrat, you mostly identified as a democratic voter. But you had a really high score on this kind of zero sum belief index. You were much more likely to vote for Donald Trump in the 2016 election than other Democrats were.

And then on the flip side, if you typically identify as a Republican but you have this high zero sum kind of ideology, You are much more likely to support redistributive policies than other Republicans. So, this kind of, framework for thinking about policy issues is something that sort of cuts across a lot of the typical party alignment. While we’re based in the U. S., Peter, you have a couple of U. S. papers and also a non U. S. paper that you wanted to tell us about today.

Peter Breadon: It’s really interesting hearing about these studies that all point to these deep rooted, social and environmental factors that shape people’s beliefs and preferences. And my studies are a little bit like that too. So I’ve put 2 studies on the list that are natural experiments.

And I think it’s fair to say the ones that you’ve both talked about are kind of quasi experimental or natural experiments as well. So, a little bit of background on this methodology and way of looking at complex systems. Natural experiments happen when, unlike a planned experiment by a researcher in a lab, it plays out in the real world.

And a change affects one group of people and not another. And you can look at whether some of the outcomes for those 2 groups diverge over time. And this means you can study things that wouldn’t be practical to do or ethical to do through a planned study. And they might be at too big a scale or over too long a time as well.

And this way of looking at things goes back a long way. So in the mid 1800s, the famous progenitor of this approach was Dr. John Snow, who looked at the cholera outbreaks in London that were killing thousands of people over many years, mapped those deaths. And found that they were clustered around this one water pump.

And although no one knew about the contamination or where cholera was coming from, that allowed them to isolate the contamination , took that pump offline. And that was kind of the beginning of this natural experiment methodology.

But my two use this approach to look at really food environments. So the first one is from the U. S., like the last study we heard about, and it’s called Sugar Sweetened Beverage Taxes and Perinatal Health, a quasi experimental study by Jackson and colleagues. And this one looked at cities in the U. S. that had a sugar tax, and then cities in the U. S. that didn’t have a sugar tax. And it found in those cities that had the tax very quickly after the introduction starting from 2015, you saw steep declines in gestational diabetes. So they looked at data about births for 5 million different births, and they found a 40 percent drop in those cities where you had the sugar tax introduced. A really striking finding. And of course, gestational diabetes has ongoing long term health risks, as well as immediate health risks for both the mother and the baby. So that’s a really big effect.

Another study was from Norway, very different setting. And in Norway, they have great data, like a lot of these Scandinavian countries where, because they have compulsory national service, Men around the age of 18 have all these measurements taken for the whole male population.

Their height, their weight, their BMI, their cognitive test scores. So they were able to link those data back to data about where fast food restaurants are located. So between 1980 and 2007, the number of fast food restaurants in Norway increased fivefold at the same time that obesity was rising very rapidly, like it has in many wealthy countries, including, of course, Australia, and they looked back and found that people who grew up closer to fast food restaurants were more likely to be overweight and the effect was very large. So they estimated it could account for about a third of all the weight gain over those three decades. So huge effect. And of course they were able to correct for all kinds of things in these rich data in terms of you know, whether parents stayed married, the education levels, the BMI of the father when, when he went through military service and had his weight collected so really, really robust and clever study, but even more interesting and troubling part of it was that, there was an effect on cognitive test scores as well. So they went down for children who grew up close to fast food outlets. And those effects were more pronounced for children who had fathers with lower educational attainment.

So this is compounding disadvantage across the generations. So, really amazing study. It’s called Swallow This childhood and adolescent exposure to fast food restaurants, BMI and cognitive ability by Abramson and colleagues, and that’s an NBR working paper.

We often think that health and particularly these things about overweight and obesity is the fault of the individual. You know, you lack discipline and willpower. But of course, obesity since 1980 in Australia, the rate has tripled, and this shows it’s not about individual, it’s about, it’s about a big change in the environment because we haven’t seen people’s preferences to be healthy or people’s willpower structurally collapse in this period.

But these studies really show it so starkly. And of course, they link to policy. We see the effect of taxing sugar sweetened beverages in changing that consumption and the effect of, you know, the kind of planning laws that you could think about imposing in terms of where fast food restaurants are and we can see really huge effects on health and they flow on, of course, to to chronic disease from obesity and expenditure on the health system and so on.

Liz Baldwin: Peter, can I ask you, as a, you know, health policy practitioner and researcher, when you’re thinking about, you know, you’re trying to, understand, say the connection between sugar sweetened beverages and health outcomes. And, you know, you, you have some excellent quite causal evidence from overseas like this from, you know, settings that might be quite different to Australia.

And then say you might have some Australian evidence that’s kind of lower quality. Maybe it’s just based on a cross sectional survey or, you know, more qualitative evidence or something like that. How do you kind of like weigh up those sort of streams of evidence, I guess, weigh up external and internal validity in a way.

Peter Breadon: really important question. I’m glad you raised it because I didn’t make the disclaimer. And none of us did that. We would never base policy decisions on a single study. But these ones are very robust and well designed using really good data. So you want to as you say, really triangulate across a range of information.

So to use these examples, there’s very strong evidence about the health impacts of drinking these sugar sweetened drinks. So we understand the mechanism through which it contributes to outcomes like diabetes. So that evidence is really clear. And then we’ve seen a growing evidence about the impact of fast food, ultra processed food consumption and sugar sweetened beverage consumption from a range of different settings that are both similar to Australia and dissimilar.

So it really has built up over time. So in this case, we see that internal and external validity you spoke about. We see a growing quantum and strength of evidence and we understand the underlying mechanism. So really, when you know why it’s happening and you know the policy works from a range of settings, that’s when we think it’s time for governments to change policy.

Liz Baldwin: While we’re talking about the evidence space writ large I’ll mention the last paper that I wanted to discuss on the Wonks’ List, which is an opinion piece written by a Harvard professor of psychology called Adam Mastrioni.

And the piece is The Rise and Fall of Peer Review, which is a pretty provocative title. He declares that peer review has completely failed. He declares the death of peer review, basically. He says, that it’s failed at its main job, which is safeguarding the quality of the research that’s published.

He cites, you know, some of the well known cases of fraud and the replication crisis in psychology and also economics and a whole range of other fields. He also refers to some pretty damning experiments where researchers have introduced deliberate really large errors into papers that they’re submitting for publication to various journals.

Those papers have gone out for peer review through the normal process. And those reviewers picked up in a lot of cases, less than a third of the major errors that the researchers had introduced. So Mastroianni argues that this means peer reviewers are failing at their kind of first job, which is gatekeeping really poor quality research.

And then on the flip side, he argues that the whole process is really costly. It’s very expensive in terms of the time that reviewers spend reviewing and that authors spend responding to those comments. It creates big delays in the publication process because it often takes, you know, years for the paper to go through rounds of revision and resubmitting, which can be a problem in kind of fast moving fields of science.

The whole process kind of gives a stamp of legitimacy to research. Like I think we’ve all probably done this before we say, Oh, well, there’s a peer reviewed article on it. And it’s sort of meant to be the debate ender. And I think this paper quite compellingly suggests that it shouldn’t be. Mastroianni’s solution to this litany of failures is pretty extreme.

He suggests that we should just do away with the whole peer review process completely. It’s unfixable. And he suggests that scientists should just move to publishing their data and their code and their papers online in a much more kind of open source format. And reviewers can comment on it in a more public way through that process.

Now, more open science is always helpful, but Tash, does this sound like a good solution to you?

Peter Breadon: I don’t know why you can’t have both a peer review and more transparency of the, of the data and the code used to manipulate it. And I do worry about the counterfactual, you know, if peer review is failing, no peer review might be even worse. I know there’s other things going on, like people putting more emphasis on replication and creating incentives for academics to do that.

And at Grattan, we rely heavily on where we can high quality systematic reviews and meta reviews. And there’s some great scoring systems online that people can use to understand whether those studies themselves that aggregate the findings for a range of studies are high quality. I think this does seem like a fairly extreme proposal to strip away peer review, but I guess on an anecdotal note, I’ve definitely been involved in peer review processes where the people providing the peer review can be relatively easily identified because they’re urging citations that all seem to have one author, which is potentially them. So look, there’s obviously flaws in all systems, but I think improving this one rather than abolishing it would be my suggestion.

Natasha Bradshaw: Going through the peer review process I’ve experienced it to be quite laborious as described, but I’m also extremely grateful for the comments and feedback that I’ve got from peer reviewers. The people that do the peer review are typically experts in the field with a vast knowledge of the area.

You know, maybe they don’t pick out all the problems, but they do pick out some, and I found that very helpful. And I wonder if you know, they put in the time and effort, if it wasn’t such a structured process, when they’re at that sort of level of expertise. So I think as Peter said, having both is helpful.

I think where you can publishing your data and your code online you know, it’s not only. helpful for review, but for giving other people a headstart on their research so they don’t waste time duplicating effort. I think that would really push things forward as well.

Liz Baldwin: Yeah, I agree. It’s such a treat when you go to research an area and you discover that someone has published some data and code online and you can, you know, do another cut of it or just dig into exactly what they’ve explored. Well, I think that’s all we have for the Wonks List 2023.

Natasha Bradshaw: Thanks for listening. If you’d like to hear more, you can find both the Wonks’ List and the main list on the Grattan website, grattan.edu.Au. Hope you have a wonderful summer and happy reading.