Friday Fodder (38/24)

Can you believe that it's already October? That means silly season is nearly here, and with it most of Australia's public holidays. So, given that most of the country will be off for Labour Day or another of the King's many birthdays on Monday, Aussienomics will too.

On to the news!

The productivity crisis rolls on

The Productivity Commission's (PC) latest update did not paint a pretty picture of the Australian economy. Productivity was down again, with Aussies making up for it by working longer hours.

One of the biggest problems with the post-pandemic economy is that most of the growth in hours work is happening in the health and social assistance industry. It, along with other "non-market" sectors, are now less productive than they were a decade ago:

If you care about real wages and living standards, that's a problem. And as the AFR's economics correspondent Michael Read recently showed, a lot of the problem can be traced to the National Disability Insurance Scheme, or NDIS:

"The rapidly expanding $49 billion National Disability Insurance Scheme is radically reshaping the national economy. It is fuelling a taxpayer-funded boom in care sector employment and underwriting a once-in-a-generation structural increase in government spending that rivals the mining boom in terms of scale.

Government funded jobs have surged – about three in four jobs filled so far this year were in the so-called 'non-market' sector – and the increase can be largely explained by a boom in healthcare employment courtesy of the NDIS.
...
Commonwealth Bank economist Harry Ottley estimates that over the past five years, health employment increased by a whopping 36 per cent, compared to just 8 per cent across all other industries."

Australia's unemployment rate is still close to all-time lows. But when most of the growth in employment is in the non-market sector (three of four new jobs over the past year), and governments at all levels are adding to net debt to pay for them, then at some point the sustainability of the entire model will have to come into question.

Qatar Airways uses the back door

In August 2022, Qatar Airways applied for permission to fly an extra 21 weekly flights into Australia on top of the 28 it already operates. Eleven months later, Transport minister Catherine King rejected the application, arguing:

"...that it wanted to protect jobs at Qantas, allow the airline to buy new planes and because it was in the national interest. Ms King – who will be called to front the Senate inquiry – has also pointed to a human rights incident at Doha Airport in 2020 as 'context'."

The whole thing was very secretive (we still don't know the real reason), and former Qantas CEO Alan Joyce – who wrote a letter objecting to the Qatar Airways application – appears to have been a central figure in influencing the government, hosting a banquet attended by Anthony Albanese and painting several of its aircraft with "Yes23" in support of the Voice referendum.

But in terms of the national interest, allowing Qatar Airways to fly would have unquestionably been good for everyone except for Qantas shareholders. The head of the ACCC, Gina Cass-Gottlieb, said allowing Qatar to fly would have reduced prices for consumers:

"She added it was hard to predict by how much it could have cut prices, but quoted Virgin Australia's estimate of 40 per cent."

So I was pleasantly surprised when it was announced this week that Qatar Airways hadn't given up and will take a different route to helping Aussie travellers:

"Qatar Airways will buy a 25 per cent stake in Virgin Australia after an agreement was reached between the Australian airline's owner, Bain Capital, and the Gulf carrier.
...
Under the proposed deal, Virgin Australia plans to launch flights from Brisbane, Melbourne, Perth and Sydney to Doha, hoping to achieve its 'measured entry' into long-haul international flying by mid-2025.

The flights would be provided under a 'wet lease' agreement with Qatar, which outlines when an airline loans an aircraft to another carrier to temporarily increase capacity."

Basically, Qatar Airways will be able to do exactly what it wanted to do, but with Virgin's name on the ticket instead of its own. Prices won't fall as much for consumers because it now has to go through a more roundabout wet leasing process and integrate its systems with Virgin, which will add to cost. But it will make travelling to the Middle East and Europe more affordable for the thousands of Aussies who fly there each year.

Most importantly, this time around it's the Foreign Investment Review Board and ACCC that have the power to approve the deal. That means it will be Treasurer Jim Chalmers, not Catherine King, who ultimately decides whether to help consumers or Qantas.

Let's just hope he has the decency to do the right thing.

Xi's big bazooka

After several months spent mired in deflation, China's government finally decided to do something about it. And markets loved it:

"The Hang Seng China Enterprises Index climbed as much as 8.5% before closing up 7.1%, a 13th straight day of gains. Property developers led the rally, with a gauge of the sector surging as much as 47%, while an index of brokerage shares jumped 35%, both record intraday moves. Mainland Chinese markets remain shut until Oct. 8 for a week-long holiday.
...
Sentiment toward equities in the world's second-biggest economy has seen a dramatic turnaround since the start of last week as the authorities unveiled a range of stimulus measures that included interest-rate cuts, freeing-up of cash for banks and liquidity support for stocks. Four major cities also eased home-buying curbs and the central bank moved to lower mortgage rates."

It was the Shanghai Composite's fastest acceleration since 1996 before it closed on Tuesday for the week-long foundation anniversary holiday. It's still below its recent 2021 highs, but needless to say, the government succeeded in lighting a fire under equities:

Commodities also rallied, including the all-important (for Australia) iron ore:

Will it last? That's the billion-dollar question, but if history is any guide then I'm inclined to say no. Stimulus might get things moving again – and as I've argued, China definitely needed some monetary easing – but when it's done this aggressively it tends to make things worse in the long-run.

That's because China needs its failing parts to die, and for the scarce resources they're using to be freed up for more productive uses. Giving them a shot of stimmy juice might help for a while but does nothing to address the structural problems in the economy, which you can be sure will come right back to the fore when the money runs out.

Peanut allergies? Blame paediatricians

The WSJ ran an interesting story on the rise of peanut allergies, which it traced to the advice doctors gave to parents:

"The American Academy of Pediatrics (AAP) wanted to respond to public concern by telling parents what they should do to protect their kids from peanut allergies. There was just one problem: Doctors didn't actually know what precautions, if any, parents should take. Rather than admit that, in the year 2000 the AAP issued a recommendation for children 0 to 3 years old and pregnant and lactating mothers to avoid all peanuts.

The AAP committee was following in the footsteps of the U.K.'s health department, which two years earlier had recommended total peanut abstinence. That recommendation was technically only for children at high risk of developing an allergy, but the AAP authors acknowledged that 'the ability to determine which infants are high-risk is imperfect.' Using the strictest interpretation, a child could qualify as high-risk if any family member had any allergy or asthma.

Many well-meaning paediatricians and parents read the recommendation and thought, 'Why take chances?'"

I did some sleuthing and according to a 2004 report by Susan Prescott of the School of Paediatrics and Child Health Research at UWA, in Australia for "children with existing sensitisations or overt allergic disease (or those deemed to be at high risk for other reasons)", it was "common clinical practice to recommend avoidance of potentially allergenic foods such as egg and milk until 12 months of age, and peanuts, nuts and shellfish until after 2-4 years of age".

So, our doctors were probably giving similar advice to those in the UK and US. The Prescott report even recommended continuing to give that advice because "this is unlikely to cause harm".

If only they knew the harm they were causing. Back to the WSJ:

"Dr. Gideon Lack, a paediatric allergist and immunologist in London, had a different view. In 2000 he was giving a lecture in Israel on allergies and asked the roughly 200 paediatricians in the audience, 'How many of you are seeing kids with a peanut allergy?' Only two or three raised their hands. Back in London, nearly every paediatrician had raised their hand to the same question.

Startled by the discrepancy, he had a eureka moment. Many Israeli infants are fed a peanut-based food called Bamba. To Lack, this was no coincidence, and he quickly assembled researchers in Tel Aviv and Jerusalem to launch a formal study. It found that Jewish children in Israel had one-tenth the rate of peanut allergies compared with Jewish children in the UK, suggesting that genetic predisposition was not responsible, as the medical establishment had assumed."

Lack has since run clinical trials, finding that early exposure leads to "an 86% reduction in peanut allergies by the time the child reached age 5 compared with children who followed the AAP recommendation".

The good news is that the official line, at least in Australia, has changed, and mothers are being encouraged to introduce peanuts before 12 months of age. The government is also funding a world-first clinical study to test the efficacy of peanut exposure during pregnancy.

The lesson here, at least for me, is that you should always get a second opinion (even if that's just checking the source of the advice!). That, and it's fortunate there were no misinformation laws in place that might have prevented people like Gideon Lack from questioning "the efficacy of preventative health measures"!

Fraud, fraud, and more fraud

The replication crisis in the social sciences (especially psychology) is, by now, well known. But as time goes on, the amount of outright fraud in medical research continues to surprise me:

"In 2016, when the US Congress unleashed a flood of new funding for Alzheimer's disease research, the National Institute on Aging (NIA) tapped veteran brain researcher Eliezer Masliah as a key leader for the effort. He took the helm at the agency’s Division of Neuroscience, whose budget—$2.6 billion in the last fiscal year—dwarfs the rest of NIA combined.
...
But over the past 2 years questions have arisen about some of Masliah's research. A Science investigation has now found that scores of his lab studies at UCSD and NIA are riddled with apparently falsified Western blots—images used to show the presence of proteins—and micrographs of brain tissue. Numerous images seem to have been inappropriately reused within and across papers, sometimes published years apart in different journals, describing divergent experimental conditions."

Masliah has published over 800 research papers, at least 132 of which contain "suspect images". His work has led to millions of dollars of neuroscience funding thrown at it for various research and clinical trials, raising "false hopes and divert[ing] patients from trials of other experimental drugs".

Incentives matter. You can manipulate data and eventually become the Dean of Stanford – and get paid millions of dollars each year along the way there – or just be another failed scientist toiling away down dead ends. Sure, you might get caught one day, but for enough morally challenged individuals the risk/reward pay-off is clearly worth it.

I just wonder with so many cases of fraud being discovered each year, how many people are out there still living the lie? And how many dollars will be wasted, and lives lost, because of those lies?

I don't pretend to know what to do about it; obviously the incentives need to be better aligned away from fraud. Perhaps a use for LLMs will be to scour research papers looking for evidence of fraud, increasing the risk for those who might want to choose that path?

AI boom or bust?

The US is all-in on large language models, if this series of charts is anything to go by:

It's not just capital but also infrastructure:

And energy; so much energy. Microsoft has even signed a deal for its own nuclear plant to power all of those new data centres:

There will no doubt be many applications for LLMs, some good, most bad. Google's new podcast generator is amazing, even if a bit cringe-worthy. Meta's partnership with Ray-Ban to make AI-powered smart glasses is very neat (not that I'd ever trust Mark Zuckerberg enough to wear his technology on my head!). OpenAI continues to expand, yesterday raising another $6.6 billion to double its valuation at $157 billion. That buys a lot of researchers, engineers and servers!

But these companies are going to need to start generating a lot of revenue or the boom will just as quickly turn to bust. I personally think a lot of this investment will fail to pay off, just like how many railroad and dot-com investors were wiped out. But that doesn't mean it won't be a huge win for consumers, especially in countries with affordable energy to keep data centre costs competitive.

Not-so fun facts

It looks like some of the problems Richard Reeves documented in his book, Of Boys and Men, also exist in Australia:

And as a follow-up on Monday's post:

"On the 1st of October all chickens in Britain must be registered with the state by law. It will be a criminal offence to keep unregistered chickens."

Britain is sclerotic by choice.

Have a great weekend.