This week in research #28

The blog's been quiet this week as I have been travelling. However, research marches on. Here's what caught my eye in research over the past week:

• Fernholz and Kramer (with ungated earlier version here) find that city population distributions for White populations in the US follow Zipf's and Gibrat's Laws from 1910 to 2020, but that Black populations only do so from the second half of the 20th Century, consistent with restricted mobility out of the South for Black populations
• Bian and Zhou find that industrial robot adoption has a significant negative effect on the internal migration inflows at the city level in China, while having little effect on internal migration outflows
• Bryson et al. (open access) find that there is little effect of head coach turnover on football (soccer) team performance, using data from the professional leagues in France, Germany, Italy, and Spain from 2000/01-2014/15

Book review: Campus Economics

Unsurprisingly, I have a strong view that everyone would be better off understanding and applying a little more economic thinking in their everyday lives, and their business and policy decision-making. In particular, recognising that there are trade-offs in every decision (TANSTAAFL - there ain't no such thing as a free lunch) would go a long way to improving the decisions that many people make. So, I had high hopes in reading Sandy Baum and Michael McPherson's book Campus Economics, which is subtitled "How economic thinking can help improve college and university decisions". And Baum and McPherson are explicit that their goal in the book is to:

...facilitate communication among groups on campus by creating a common vocabulary and encouraging modes of thinking that allow participants to better see other viewpoints and grapple with the trade-offs involved in making sound decisions.

Unfortunately, my high hopes for this book would go mostly unfulfilled. Much of the book is devoted to a descriptive presentation of data on college and universities, with a particular focus on enrolment and financial data. That sort of data has its place, and is interesting in some ways. However, the book didn't add a whole lot of value beyond the data, other than posing some questions to consider in various decision-making situations, including decisions related to faculty compensation, budget cuts, tuition pricing, college endowments. Now, the tertiary education sector in the US is quite diverse, with public, private not-for-profit, and private for-profit institutions, all with different goals and constraints, but nevertheless there are common elements to all of them and Baum and McPherson don't go very deep on any of these areas of decision-making. What we are left with is a once-over-lightly approach to a number of topics, which isn't very satisfying.

The book does have some positive aspects. For those unfamiliar with the US tertiary education context, it does provide a good primer, and may be useful for that purpose. Baum and McPherson present economic concepts close to the beginning of the book, but mostly avoid using the technical terms (like opportunity cost) thereafter, but I feel like the book would have been better if the technical terms (occasionally reminding readers of their meaning) had made more frequent appearances. In discussing college endowments, Baum and McPherson did point out one thing that many people should have a better understanding of, which is fungibility:

However, if the business curriculum is funded partly from the endowment gift and partly from general university funds, there is nothing to stop the university from reducing the general funds going to business and directing that money to hiring faculty in the arts. Because very few activities at universities are fully funded by endowment gifts, even restricted endowments are more "fungible" than they look.

Indeed, there are many occasions where funds that are earmarked for one purpose may nevertheless lead to no additional spending on that purpose, and we should always be conscious of that possibility.

Overall, despite those positive aspects, the people who would get the most out of this book are those who are interested in campus finances in higher education in the US. However, even then, the data are likely to get dated reasonably fast, even if the underlying concepts remain relevant. Beyond that, I don't think that the book provides enough depth to sustain a wider readership.

This week in research #27

Here's what caught my eye in research over the past week:

• Whitelaw and Branson (open access) estimate the effects of pandemic-related closures on the academic performance trajectories of undergraduate students at a university in South Africa, and find performance gains in 2020, that were reversed in 2021, and that the achievement gap between students from differing socio-economic backgrounds increased
• Similarly, Su et al. (open access) look at the impact of zero-COVID policies on the academic performance of primary and secondary students in China, and find that the stringency of the zero-COVID policy is associated with significantly better mathematics performance for boys, while having no effects on girls
• Schläpfer finds that football (soccer) players from a cultural background that places a higher value on revenge are more likely to retaliate for a foul during a game but are not more likely to commit fouls overall, using data from nine football leagues over the period from 2016 to 2019
• Jeong and Lee (open access) introduce a spatial autoregressive hurdle model for nonnegative origin-destination flows (pretty technical, but interesting to those of us who may want to create econometric models or migration or trade flows that take into account spatial spillovers
• Niki (open access) uses TIMSS data from Japan to show that a reduction in instruction time due to the revision of curriculum guidelines in Japan reduced student test scores and motivation (no surprises there?)
• Anand and Kahn (with ungated earlier version here) find that teenage girls who observe a friend or older sibling’s teen pregnancy are less likely to have unprotected sex and have fewer sexual partners in the year following the end of the teen pregnancy
• Celhay, Depetris-Chauvin, and Riquelme (with ungated earlier version here) find that the 2011 nationwide student strike in Chile led to an average increase of 2.7% in teenage pregnancies (I guess students found something to do in their spare time during the strike?)

How a good growing season in Australia made New Zealand avocado growers worse off

RNZ reported earlier this week:

It had been tough going, New Zealand Avocado chief executive Brad Siebert said, but the new season which is just getting started looks good...

The glut of avocados last season was in part because Australia, a key market for NZ exports, had grown more of its own.

"Over the last decade Australia has taken around 85 percent of our exports but in the last three years that's dropped below 50 percent."...

Many growers have had two seasons in a row without any profit, he said.

"It's been really tough so there is a bit riding on this season."

According to New Zealand Avocado, Australia is New Zealand's largest export market for avocados, taking about 80 percent of our exports. A good growing season for avocados in Australia will therefore make life difficult for New Zealand avocado growers.

To see why, first consider the Australian avocado market, as shown in the diagram below. The domestic supply of avocados in Australia is S0, and demand is D. If there was no trade, the market would operate in equilibrium, where supply meets demand. The equilibrium quantity would be Q0, and the equilibrium price would be PD. However, Australia is an importer of avocados, which means that the domestic equilibrium price in Australia is higher than the world price (PW) - Australia is less efficient at producing avocados than the rest of the world, because they do so at a higher price than the world market. Since Australia can trade for avocados on the world market, and Australian consumers have the choice, they would not pay more than PW for avocados. At the world price PW, the domestic suppliers of avocados are willing to supply just QS avocados. However, domestic consumers demand Q1 avocados. The difference between QD and QS is the quantity of avocados imported (equal to M0). So, the effective supply of avocados to the Australian market is shown by the red line, where the domestic sellers sell up to QS avocados, and then the rest of the supply comes from the world market at the price of PW. 

Now consider what happens when there is a good growing season in Australia. This increases the domestic supply of avocados from S0 to S1. Now, at the price PW, Australian avocado growers are willing to supply Q0 avocados [*], while domestic consumers still demand QD avocados. The difference between QD and Q0 is the quantity of avocados imported now (equal to M1, and smaller than M0). The quantity of avocados imported into Australia declines.

Obviously, since New Zealand is a major contributor to Australian imports of avocados, this will make New Zealand avocado growers worse off, at least initially. To see why, consumer the New Zealand avocado market, shown in the diagram below. In this case, New Zealand is an exporting country, so the world price (PW) is higher than the domestic price (PA) - New Zealand is more efficient at producing avocados than the rest of the world, because they can do so at a lower price than the world market. Avocado growers in New Zealand would prefer to sell avocados to the world market for PW rather than locally at the price of PA. So, New Zealand consumers must also pay the world price for avocados. At that price, the domestic suppliers of avocados are willing to supply QSA avocados, but domestic consumers are only willing to buy QDA avocados. The difference between QSA and QDA is the quantity of avocados exported (equal to XA). The effective demand for New Zealand avocados, including exports, is shown by the blue line, where the domestic consumers buy up to QDA, and then the rest of the demand comes from the world market at the price of PW.

Now consider what happens when Australia reduces imports of avocados. In the short run, New Zealand can't easily divert those exports to other destinations in the world market. So, those avocados would now be supplied to the domestic market instead. So, instead of XA exports, there can only be XB exports. [**] If domestic consumers only bought QDA avocados, and growers could only export XB avocados, then not all of the growers' avocados are being sold. The growers produced QSA avocados, but are only able to sell QA (which is QDA plus XB). To sell those leftover avocados, the growers must go back to the New Zealand market. This is shown by the dotted blue line DC (incl. exports). Equilibrium in this market now occurs where that new demand curve meets the supply curve, which is at the quantity QC. The domestic price of avocados would increase to PC.

Now, to show that the domestic producers are made worse off, we need to consider the areas of economic welfare. Before the decrease in exports, the consumer surplus (the gains to domestic avocado consumers) is the area ABPW, the producer surplus (the gains to domestic avocado growers) is the area PWCE), and total welfare (the sum of consumer surplus and producer surplus, or the gains to society overall) would be the area ABCE. After the decrease in exports, the consumer surplus increases to the area AFPC. Domestic consumers are made better off. The producer surplus decreases to the areas PCGE and HJGF combined [***]. Domestic growers are made worse off. Total welfare decreases to the areas AFGE and HJGF combined. Society as a whole is worse off. However, the key point of relevance to this post is that the growers are worse off.

Now, in the long run, New Zealand growers might be able to find other export markets, so that when Australian demand for imported avocados decreases, New Zealand isn't as badly affected. As the RNZ article notes:

Siebert said the industry had worked hard to diversify markets and is exporting to more countries this season than ever before.

"We're going back into Canada and North America, we haven't exported there in a number of years, we're also going into Asia.

That way at least, less avocados would need to be dumped on the domestic market in future, making the avocado growers better off (but, sadly, New Zealand avocado consumers would be made worse off). 

*****

[*] For simplicity in the diagram, I've assumed that the quantity supplied at the world price after the supply increase is exactly equal to Q0. It wouldn't be exactly that, but it makes the diagram a bit simpler.

[**] Again, I've assumed that the quantity of New Zealand exports is reduced by half. This just makes the diagram a bit simpler.

[***] This requires a bit of explanation. The sales of QE avocados to the domestic market generates producer surplus equal to the area PCFKE. Where does QE come from? It is the initial sales to domestic consumers of QDA, plus the extra sales from QA to QC (shifted over to the left so that the domestic sales are merged together - notice this just moves the two parts of the domestic demand curve, D and DC, together). Then the sales of export avocados (form QE to QC) generates additional producer surplus equal to the area HJGK. Combining those two areas gives PCGE + HJGF.

Book review: The Moral Economy

A famous case of incentives going wrong happened in a childcare centre in Haifa, and was reported in this study (ungated here) by Uri Gneezy and Aldo Rustichini (also discussed in Uri Gneezy's book with John List, The Why Axis, which I reviewed here). The problem was that many parents picked up their children late. In the experiment, late parents were made to pay a modest fine for their lateness. This should have created an incentive for parents to pick up their children on time. Instead, it resulted in more late pick-ups.

This incentive failure is one of the motivating examples in Sam Bowles' 2016 book The Moral Economy, which I just finished reading. Bowles draws a distinction between the economic incentives that decision-makers face, and the moral and social norms that have been established. This is not a new distinction. In fact, Steven Levitt and Stephen Dubner drew a distinction between economic incentives (which are predominantly monetary), moral incentives (based on what the decision-maker believes is right and wrong), and social incentives (based on what other people perceive is right and wrong), in their famous book Freakonomics. However, in his book Bowles describes how these different incentives may be at odds with each other. In the case of the Haifa childcare experiment, the economic incentive would have led to fewer late pickups, but it changed the norms of picking up on time, reducing the moral and social incentives against late pickups. The combined result: more late pickups.

Bowles argues that incentives have both direct and indirect effects on behaviour. The direct effect is based on the economic incentive, but the indirect effect works through decision-makers experienced values or social preferences. Sometimes, there is no indirect effect. The economic incentive and social preferences are independent of each other. Bowles refers to this as separability. However, when there is an indirect effect of the economic incentive on social preferences, and:

When the indirect effect is negative, meaning that the total effect falls short of the direct effect, then incentives and social preferences are substitutes (or are "sub-additive" or are said to exhibit "negative synergy" or "crowding out")...

Where the indirect effect is negative and large enough to offset the direct effect of the incentive, we have the attention-riveting cases in which incentives backfire, that is, they have the opposite of the intended effect, which I term "strong crowding out"...

Where the indirect effect is positive, we have crowding in, that is, synergy between the two effects: then incentives and social preferences are complements rather than substitutes, and are sometimes termed "superadditive".

A good portion of the middle of the book is devoted to outlining a number of experimental (both lab experiments and field experiments) studies that illustrate these four effects. Bowles also distinguishes between marginal crowding out (where the economic incentive has an indirect effect, and the size of the indirect effect scales with the size of the incentive) and categorical crowding out (where an indirect effect occurs just because the economic incentive exists, regardless of the size of the economic incentive). A combination of both marginal and categorical crowding out is possible, as are marginal and categorical crowding in.

If all of that sounds very complicated, you are not wrong. Bowles does his best to make the book interesting and engaging, but the underlying material is quite technical (albeit not particularly mathematical), and so the key messages from the book are likely to pass by the general reader. Economists will be more likely to understand the material, but perhaps less likely to fully accept what they mean for incentives.

My takeaways from the book were that economic incentives may erode social motivations, and that means that economic incentives may not be as effective as we think they are, and in fact in some circumstances may have the opposite of the intended effect. Of course, these are not new conclusions, as Gneezy's study (and others) have highlighted this much earlier. This book simply compiles more of the evidence in one place, and provides a framework for understanding it.

Where the book falls short, though, is the policy prescription. If incentives don't work as well as intended, or may have the opposite of the intended effect, can policy-makers (or others) determine in advance what the effects will be? How can we know the circumstances under which economic incentives will crowd out social preferences, and especially how can we know if the economic incentives will be counter-productive? Here, the framework itself is little help, and unfortunately, Bowles has little guidance. In the conclusion, he writes that:

I do not know whether an approach to constitutions, incentives and sanctions adequate to this challenge can be developed. But we have little choice but to try.

That may be a fair assessment, but is a bit of a let-down. Hopefully though, this indicates the start of a programme of research by others following in Bowles' footsteps, to provide more guidance on how incentives fail. Those current and future researchers may be the best audience for this book.

The 'mighty girl effect' may only kick in when daughters reach school age

You may have heard of the 'mighty girl effect' (also called the 'eldest daughter effect') - the idea that fathers whose eldest child is a daughter are less likely to support traditional gender norms, and have more progressive views. There are several studies that support the existence of this effect (see here or here for examples). However, less studied is when this effect occurs. Does the birth of a daughter have an immediate impact, or does it take time for fathers to change their views? And if it takes time for father's views to change, how long does it take?

This 2019 article by Mireia Borrell-Porta, Joan Costa-Font, and Julia Philipp (all London School of Economics and Political Science), published in the journal Oxford Economic Papers (open access), provides an initial answer. They used data from the British Household Panel Survey waves between 1991 and 2012, a sample of nearly 28,000 observations of over 11,000 parents (about 44 percent men). Their key measure was agreement with the statement "a husband’s job is to earn money; a wife’s job is to look after the home and family", initially measured on a five-point scale ranging from "strongly agree" to "strongly disagree", but in most analyses they use a binary version of the variable, set equal to one where the respondent strongly agreed, agreed, or neither agreed nor disagreed with the statement (thereby demonstrating some level of support for traditional gender roles).

Borrell-Porta et al. take advantage of the fact that the gender of a child is essentially random, and compare parents with at least one daughter in the household with those with no daughters. They then extend that analysis (which is similar to previous research) to consider the age of the oldest daughter (in three categories: 0 to 5 years; 6 to 10 years; and 11 to 18 years). The first set of results are well summarised in the simplest analysis, presented in Figure 1 of the paper:

Fathers with daughters are less likely to support traditional gender roles, but the results are less clear for mothers. So far, nothing so different from earlier work in this area. In the full analysis, separating daughters by age, Borrell-Porta et al. find that:

...fathers’ probability to support traditional gender norms declines by approximately three percentage points (8% change) when parenting primary school-aged daughters and by four percentage points (11% change) when parenting secondary school-aged daughters. In contrast, the effect on mothers’ attitudes is smaller and generally not statistically significant.

All of that is based on self-reported responses to the question, so Borrell-Porta et al. look deeper for behavioural change, specifically looking at whether fathers of daughters are less likely to be in a couple that follows a 'male breadwinner norm' (with the father working, and the mother not working). They find that:

Parenting pre-school daughters is associated with a higher probability to behave traditionally. However, parenting primary and secondary school-age daughters is associated with a lower likelihood to follow a traditional male breadwinner norm in which the man works and the woman does not work, and this result holds both cross-sectionally and longitudinally. In terms of effect size, FEs estimates... indicate that parenting daughters aged six to 10 reduces the probability of a traditional gender division of work by seven percentage points, and parenting daughters aged 11 or older reduces that probability by five percentage points. Compared to the baseline probability of following a traditional norm for those without daughters of 20.3%, this is a sizeable reduction of 36% and 25%, respectively.

So, being father to a daughter not only appears to change attitudes, but also behaviour, but only when those daughters reach school-age. Borrell-Porta et al. note that these results are consistent with exposure theory (which says that men develop or change their understanding of women's place in society when exposed to situations that make them more sympathetic - something that mothers would have already experienced, but fathers experience through their daughters), as well as identity theory (which says that the child's wellbeing enters into the parent's utility function - that is, the parent feels better off when the child is better off). Borrell-Porta et al. aren't able to tease apart those possible mechanisms underlying the results.

I find these studies interesting, but I think they raise as many questions as they answer. Fathers have had daughters for millenia. If each generation of fathers became less likely to support 'traditional' gender norms than the previous generation, by the amounts that these studies find, then the 'traditional' gender norms would have disappeared long ago. What is really missing here is an answer to the question of, why now? Is there something about modern times that facilitates this change? Are there certain pre-conditions that need to be in place before daughters can have an appreciable impact on the attitudes and behaviours of their fathers? Do these results hold across cultures that are less progressive than the UK and the US (where the studies that I have seen have been based)? These are all questions that would be interesting to answer, and give us a better understanding of how daughters contribute to the breakdown of fathers' support for traditional gender norms.

This week in research #26

Here's what caught my eye in research over the past week:

• Habla et al. (open access) ask research participants in Sweden how they would like self‐driving cars to be programmed for dilemma situations (essentially choosing between harming pedestrians or passengers), and find that at the margin, the average respondent values the lives of passengers and pedestrians equally when both groups are homogeneous and no group is to blame for the dilemma, but that they value the lives of passengers more when the pedestrians violate a social norm, and less when the pedestrians are children
• Brown et al. (with ungated earlier version here) conduct a new meta-analysis of 607 loss aversion estimates, finding that people are loss averse on average, but somewhat less loss averse than assumed in previous literature (the debate continues, see here and here for more)
• Franceschi et al. (open access) provide a systematic review of the literature on the determinants of the value of football (soccer) players

Why many consumers will switch from orange juice to mandarin juice, and be paying more for it

The Financial Times reported last week (probably paywalled, in which case try this article from The Conversation on the same topic):

Orange juice prices have soared to record highs, driven by bad weather and disease in Brazil, the world’s largest exporter, prompting manufacturers to explore whether they can use mandarins instead to make the drink...

The long-term solution to a dearth of oranges, according to [president of the International Fruit and Vegetable Juice Association, Kees] Cools, might be to make orange juice from mandarins, whose trees are more resilient to climate change in growing regions...

The industry is already experimenting. In Japan, which normally imports 90 per cent of its orange juice, mostly from fruit grown in Brazil, the supply crunch has been exacerbated by a weak yen, pushing up import costs further. Seven & i Holdings, the owner of supermarket chain 7-Eleven, has turned to the country’s domestic supply of mandarins, launching a mandarin and orange juice product.

The effects of poor weather and citrus greening disease in Brazil on the markets for oranges and mandarins can be easily analysed using the supply and demand model (from my ECONS101 or ECONS102 classes). The effect on the market for oranges is shown in the diagram below. The market was initially in equilibrium, where demand D0 meets supply S0, with a price of P0 and a quantity of oranges traded of Q0. Bad weather and disease reduce the orange harvest, decreasing supply to S1. This increases the equilibrium price of oranges to P1, and reduces the quantity of oranges traded to Q1.

Now consider the market for mandarins. Mandarins are a substitute for oranges in the production of citrus juice. Since mandarins are now relatively cheaper than oranges, some juice producers will switch from making juice from oranges only, to a blend of oranges and mandarins. The effect is shown in the diagram below, where the mandarin market is initially in equilibrium with a price of PA, and a quantity of mandarins traded of QA. Producers switching to a blend of oranges and mandarins increases the demand for mandarins from DA to DB, increasing the equilibrium price of mandarins from PA to PB, and increasing the quantity of mandarins traded from QA to QB.

All of this might be bad news for orange and mandarin lovers, since the price of their favourite citrus fruits are likely to rise. Luckily, some of us have our own mandarin trees (and in fact, my family's tiny mandarin trees have been so laden with fruit this year, we've been giving them away!).

The consequences of low-quality alcohol licensing data

I've been meaning to post on this for a little while now. Back in April, Eric Crampton pointed to Police and the Medical Officer of Health using incorrect data on the number of licences in Wellington as part of their submissions opposing alcohol licences. Sadly, for those of us who know about the alcohol licensing data in New Zealand, this sort of outcome won't come as a surprise.

Current alcohol licence data is available for free from the Ministry of Justice website. However, the quality of that data is somewhat questionable. It doesn't take long to realise that there are problems. For example, I downloaded the latest data, and looked at Waikato District (which I know reasonably well, as I am a Commissioner of the District Licensing Committee there). Looking only at on-licences, McGinty's bar in Huntly appears in the list twice. Fortunately, it looks like that's the only duplicate in the Waikato data for on-licences, but it would not surprise me at all to find out that there are many duplicates for Wellington. So, simply getting the free data and counting the number of licences in that data is going to over-state the number of licences.

I think that this particular issue arises because, when a new licence is granted to new owners of an existing premises, the previous licence remains in the dataset until it expires. Once you know that it is an issue, identifying duplicates and removing them is straightforward (although not helped by the dataset having only street names and not street numbers in the address fields).

However, there is a bigger issue with the dataset. It is only updated when the local council updates the Alcohol Regulatory and Licensing Authority (ARLA, which is part of the Ministry of Justice). If the local council doesn't send updates very regularly (or at all), then the dataset can quickly get woefully out of date.

So, as one example, according to the latest dataset, there is only one licensed premise in the entirety of the South Taranaki District, and only two in the entirety of Waitaki District. Obviously, the data are incorrect there, and almost certainly because those districts are not updating ARLA when new licences are issued, or existing licences are renewed. So, the licences drop out of the dataset as they 'expire', when they are really being renewed and ARLA just doesn't know.

This is essentially the reason that I have hit 'pause' on research on alcohol outlets across the country as a whole (while maintaining some research in Hamilton City and South Auckland, where I know the quality of data is high). We really need a more reliable source of data than the dataset that ARLA maintains and makes available through the Ministry of Justice website. To be clear, I don't think it's ARLA's fault here at all. I strongly suspect that the dataset is sub-par because ARLA aren't being supported by the councils actually updating them, as is mandated by the Sale and Supply of Alcohol Act. A large part of the problem is probably that there doesn't appear to be any sanction against councils that fail to update ARLA.

One day, when I have a lot of spare resources (especially time, but also funding), we may be able to pull together a reasonable dataset that could avoid all of these problems. It would likely require a whole bunch of LGOIMA requests of councils that don't make their licensing decisions available online (again, as requires by the Sale and Supply of Alcohol Act), and a lot of effort. Some day.

In the meantime, we should spare a thought for those that want to use data on the number of alcohol licences (in Wellington, or anywhere else). The data that they are being asked to work with is not fit for purpose.

Michael Ryan on whether fighting inflation always leads to recession

We've just finished teaching the macroeconomics section of ECONS101 for this trimester (which brings our teaching to a close). The last week covers the Phillips Curve - the empirically-observed short-run trade-off between inflation and unemployment. The implication of the Phillips Curve is that if the government wants to reduce inflation, it can do so only at the cost of higher unemployment. And if the government wants to reduce unemployment, it can do so only at the cost of higher inflation.

My colleague Michael Ryan wrote on The Conversation recently on the topic of whether fighting inflation leads to recession. He wrote:

But are reductions in inflation inextricably linked to recessions?

New Zealand’s own economic history, it turns out, can give some guidance on this, and point to the risk factors within the country’s economic outlook...

Since 1961, New Zealand has experienced eight falls in inflation (disinflations) of four percentage points or more. (Disinflation refers to when inflation drops but remains positive, while “deflation” occurs when the inflation rate falls below zero).

This four percentage point drop is required for New Zealand’s inflation to reach the Reserve Bank’s target of 1-3%, down from the 7.3% recorded in the third quarter of 2022...

Then, after looking at New Zealand's history of periods of disinflation since 1960, he concludes that:

The message is a positive one: a fall in inflation does not necessarily have to be associated with a recession.

That was a bit of a relief to me, given that I wrote at the end of 2022 (also in The Conversation) that the Phillips Curve relationship is not causal, but that nevertheless:

...we can probably expect unemployment to move upwards as the Reserve Bank’s inflation battle continues. Not because lower inflation causes higher unemployment, but because worker and consumer expectations take time to reflect the likelihood of lower future inflation due to the Reserve Bank’s actions.

And since workers negotiate only infrequently with employers, there is an inevitable lag between inflation expectations changing and this being reflected in wages. Alas, for ordinary households, there is no quick and easy way out of this situation.

It is good that Michael Ryan and I are not inconsistent with each other! In theory at least, when the Reserve Bank manages to reduce inflation and unemployment is not negatively affected (that is, the economy doesn't enter recession), it's because inflation expectations have adjusted quickly. That is not always the case.