Of course we don't know what the AI algorithms do, that's the point of using them

We've yet another instance of the engineers' fallacy concerning the economy. Which is that we can understand it as a detailed level and therefore guide and even plan it. This is, as we all recognise, that Socialist Calculation debate all over again, the answer to which is that we can't, as in Hayek's The Pretence of Knowledge.

Here the fallacy comes from this insistence that we've got to know what all those algorithms are doing. For if we don't then we don't know what they're doing, do we? The answer to which is yes, quite, that's the point.

“In some ways we’ve lost agency. When programs pass into code and code passes into algorithms and then algorithms start to create new algorithms, it gets farther and farther from human agency. Software is released into a code universe which no one can fully understand.”

Entirely so, that's why we're using the algos. That real world out there is complex. Too much so for us to fully understand in detail. That's why planning doesn't work - the centre can never gain enough information in useful time to be able to process it. Which is exactly why we're using the algos, isn't it, just as we use markets to guide the economy.

The insistence that we've got to examine each and every algorithm, in order to see what it's doing, so that we can direct it, is exactly that socialist - or here that desire of engineers to, well, engineer everything - planning fallacy. We'll not be able to do it, doing it would be contrary to the very reason we're using the software in the first place.

Rather better for us to get on with what we do in that economy therefore. For the idea that we'll ultimately know it all is a delusion. 

Emissions testing fakery hasn't cost drivers €150 billion, nonsense

That the car manufacturers have been, to put it most delicately, specially preparing cars that go through the emissions tests is true. It is possible, but by no means certain, that this has cost drivers some amount of money. That the number is €150 billion is definitively untrue:

Drivers in Europe have paid €150bn more on fuel than they would have if their vehicles had performed as well on-the-road as in official laboratory-based tests, according to a new report.

Car companies have legally gamed official tests of fuel economy for many years by, for example, using very hard tyres during tests or taking out equipment to make cars lighter. The gap between test and actual performance has soared from 9% in 2000 to 42% today.

Analysts at research and campaign group Transport & Environment have now calculated that this difference cost motorists in Europe €150bn (£136bn) in extra fuel between 2000 and 2017. UK drivers paid €3.5bn more in 2017 alone, and a total of €24bn since 2000.

Consider the specific problem with diesel. It is possible to make a low NoX emissions diesel with good mpg, which is expensive. It is possible to make a good mpg cheap diesel which has higher NoX emissions. What is not available - physics - is a low NoX, high mpg, low cost, diesel. 

We're thus in swings and roundabouts territory, costs and benefits. Meeting the varied EU emissions tests costs money. So too with mileage. It may well be that consumers have paid more for more fuel. But how much less have they paid for cheaper engines, cheaper cars?  

If there are c. 15 million new car registrations a year, worth €20,000 each, then the market is €300 billion a year. Over 20 years, give or take, those emissions dodging cars only have to be 2.5% cheaper for it to be a wash for the consumer. And if it's 3% cheaper then they're gaining. Note that we're not insisting those numbers are correct, only ballpark for logical purposes.

In fact, we'd rather like to know exactly what those relative costs are, the size of the market, for that's the manner in which we can work out whether those mpg and emissions standards are actually worth it. Do they cost consumers more, in capital costs, than the fuel savings? Interesting question, no? What is gained on the swings of lower fuel consumption is lost on the roundabouts of higher purchase costs? 

Yes, excellent work there, now, what's the net effect?

We've a finding that air pollution damages the human intellect. Significant such is estimated to limit brain power equivalent to a year of education. No, no jokes about how being free of a year of British propagandistic education makes you smarter.

The important thing we need to know being not what is the effect of pollution, but what is the net effect of pollution having been created? 

Air pollution causes a “huge” reduction in intelligence, according to new research, indicating that the damage to society of toxic air is far deeper than the well-known impacts on physical health.

The research was conducted in China but is relevant across the world, with 95% of the global population breathing unsafe air. It found that high pollution levels led to significant drops in test scores in language and arithmetic, with the average impact equivalent to having lost a year of the person’s education.

“Polluted air can cause everyone to reduce their level of education by one year, which is huge,” said Xi Chen at Yale School of Public Health in the US, a member of the research team. “But we know the effect is worse for the elderly, especially those over 64, and for men, and for those with low education. If we calculate [the loss] for those, it may be a few years of education.”

We do rather think that, at times at least, the creation of pollution is net beneficial. The production or processing of food that enables people to not be stunted or brain damaged from malnutrition we might think adds to human intelligence. The creation of an economic surplus so that we've got an education system perhaps.

We've even empirical evidence. As nations have joined the soot spewing lower middle class we've seen IQ measurements rise, that Flynn Effect. That is, we know that, up to a point at least, more pollution  - as a result of what causes the pollution - is net beneficial.

The point here being that there are always costs as well as benefits to something and it's the net position we want to know about, not just the one or the other. The truth coming from this being that many a poor place out there would benefit from more pollution - or at least, the costs of the pollution would be less than the benefit coming from the activities which create the pollution. Pollution's a cost of people making a living, making a living is a benefit. Net is what matters, not just the cost of the pollution.

And yet we're thought to be the odd, delusional, ones

One of those less than delightful bits of cognitive dissonance on display here in The Guardian. The general view of that newspaper being that every problem in the world will be solved by the application of more government to whatever it is. We're then presented with the list of things that government really should be working upon:

Meanwhile, many people will wonder why there is such little interest in an array of massively important subjects. Labour is bound to sound off about the condition of the country more than the Tories, but even so, on both sides the no-shows will be obvious. Even after a summer that has flagged the realities of a warming planet in no end of vivid ways, there will probably be precious little attention paid to climate change. No one is likely to speak convincingly about how to radically change a benefits system that is broken beyond repair; nor about a schools system that is increasingly unfit for the future. The profound challenges presented by an ageing population will be skirted over, at best. The fact that councils are now colliding with bankruptcy might be mentioned, but is unlikely to lead to any deep discussion of how to change our system of local government.

Every economist on the planet has been insisting that a carbon tax is the correct and complete solution to climate change. The government's own report even stated so, the Stern Review going on to say that we should not have plans, targets, schemes, but a carbon tax. We've got plans, targets, schemes, and no carbon tax.

The benefits system is, by definition, a government incompetence and has been for many centuries now, at least since QE I. The schools are and have been government since the absorption (perhaps nationalisation) of the Church schools in the 19th century. Demography moves pretty slowly, we've known of the perils of an ageing population for decades now as we all age, that one day at a time, toward the grave. Local councils actually are government and obviously always have been.

So, our five major problems as elucidated by those who continually campaign for ever more government. Those five major problems entirely cocked by government itself, four of them for at least a century.

And yet we're the delusional ones for arguing that markets are, a little more than just occasionally, a pretty cool solution to our varied problems? 

Which is where the cognitive dissonance comes in. Given the actual track record of government in dealing with the outlined problems why do they think that government is the solution?

Why? No, really, why?

There's an important question that needs to be answered here:

Theresa May's housing adviser backs a controversial campaign to force landowners to offer huge discounts on the price of their land, it can be revealed.

Toby Lloyd, who was hired by Mrs May in April, called for an overhaul of compulsory purchase laws months before his appointment to Downing Street.

Writing on the website of Shelter, Mr Lloyd, then head of policy at the housing and homelessness charity, said the government should be able to buy up land at its "true market value", rather than current rates, which generally include a speculative uplift based on planning permission the site could gain for future development.

No, not that silliness about option value upon future development potential. What does anyone think market prices are if they're not an attempt to look into the future? 

Rather, hiring someone from Shelter? 

Sure, we can take Mancur Olson to be right here,democratic politics is simply about who gets to dip their ladle into the gravy. Thus pressure groups like Shelter are just the front men for one specific interest group - presumably those who would gain jobs, prominence and gongs from there being more money for state subsidised house building.

And yes, Spads and the like are going to be hired from those interest groups as Mancur points out. But hiring one from the enemy camp? 

That's the question that needs answering, why is a Tory government hiring someone from Shelter? It's as absurd as Momentum asking one of us to advise upon utility renationalisation.

What on Earth are they doing there at Number 10?

Venezuela Campaign: Sanctions

Supporters of the Venezuelan regime promote the idea that sanctions on Venezuela are responsible for its current economic woes, and that if the US and EU would withdraw these sanctions then everything would return to normal. This belief is categorically untrue and those who advocate it are at best disingenuous. This article will examine sanctions on Venezuela, and the reasons why they have had nothing to do with Venezuela’s economic catastrophe.

The EU began sanctions against Venezuela in November 2017 with an arms embargo, and escalated its sanctions with asset freezes and travel bans on seven Venezuelan officials in January and eleven in June 2018. To argue that an arms embargo and targeted sanctions on 18 people are responsible for a Venezuelan economic crisis which began in 2012 is absurd. An arms embargo does not prevent toilet paper and medicine reaching Venezuela.

The US sanctions started with an Act of Congress in 2014, followed by an executive order in March 2015, both solely concerned with assets freezes and limits on US entry of Venezuelan officials involved in violence and illegality.

Further US sanctions in August 2017, March 2018, and May 2018 forbid US citizens from buying Venezuelan government or state oil company debt, or buying the (largely imaginary) Petro cryptocurrency. Notably, the sanctions do not ban the import of goods into Venezuela nor the import of Venezuelan oil or other goods into the US. Their economic effect is minuscule, notably because no-one in their right mind would buy Venezuelan debt now, as the country is completely bankrupt. The executive order banning the purchase of debt outright was only issued a few months ago. Even if it had some minor economic effect, how could it have caused a deep crisis that was already spiraling out of control in 2015?

Blaming the US and EU for sanctioning Venezuela and causing its collapse is fantasy. Not only did the US and EU only sanction Venezuela very recently, but most sanctions affect only corrupt and violent regime members, with the few economic ones having no real effect.  If the US wished to cause serious economic harm to Venezuela it would stop buying its oil. Regardless, that option will soon cease to exist, as the Venezuelan regime’s incompetence will shortly succeed destroying its entire oil industry. There are many factors in Venezuela’s downfall, that much is strikingly obvious to any observer: corruption, oil dependence, Dutch disease, and irresponsible economic policies. But sanctions are not one.

More information on the Venezuela Campaign can be found on their website. 

An entirely lovely conjunction of stories about housing

The Guardian wants to tell us that freedom from the planning permission system is a very bad idea indeed. The ability to convert office space into flats, without the encumbrance of bureaucratic permission, leads to, well, leads to:

The conversion appears to have gone ahead as planned, which means each of the six upper floors is now made up of 10 self-contained studio flats. So that’s 60 flats. The architect’s drawings describe 18 of the studios as “singles” and 42 as “doubles”.

According to the plans, the smallest singles are just 13 sq metres – that’s a room just a fraction smaller than 12ft by 12ft – while the smallest doubles are 14.7 sq metres. Yet the government’s own space standard – known as the “nationally described space standard” – states that the minimum floor area for a new one-bedroom one-person home (including conversions) is 37 sq metres, and for a one-bed two person home it is 50 sq metres. While these minimum sizes are not compulsory, they do apply in London, but only to schemes that go through the planning system.

Apparently the provision of 60 housing units by side swerving the planning system is an outrage. We would say that having to side swerve the system to produce the required housing units tells us we need to reform, or ditch, the planning system. 

We have indeed pointed this out before. Even insisted that such minimal space is part of the solution.

What we find truly fun though is the second story covered in that same column. Or perhaps the conjunction of the two. For we get this: 

The research by LABC Warranty, which provides warranties for new-build homes, used data from property websites Rightmove and Zoopla to analyse house sizes within a five-mile radius of 20 UK cities.

It said that in Sheffield, the average floor space of a privately owned home was 61.2 sq metres.

The researchers collected data including the size of the living room, kitchen, master bedroom and bathroom, as well as determining the average number of bedrooms and whether the property had access to a garden. However, it appears the analysis did not include certain areas, such as a hallway or staircase, which means it may underestimate the true picture.

The three cities with the next-smallest average house sizes were Southampton (64.9 sq metres), Bristol (65 sq metres) and Glasgow (65.2 sq metres). Meanwhile, the figures for London, Manchester and Birmingham were 65.7 sq metres, 67.2 sq metres and 69.9 sq metres respectively.

That's the result of what we get from people obeying the planning system. Rabbit hutches which are, by far, the smallest average sizes for housing in Europe. Actually, one of us inhabits a rural cottage on the continent. Originally built for a landless farm labourer, never known to be one of the richer sections of society. It's a good 50% larger than those average sizes above. The British planning system produces housing smaller than that available to a Portuguese peasant near a century ago.

This is not a recommendation for that British planning system now, is it? For that British planning system produces not enough and small housing. Where it's possible to dodge it housing most certainly gets provided in volume. All we need to do now is to obliterate the restrictions against building housing people actually want to live in and we'll be done.

We even have history as a guide for us. Those 1930s semis and detacheds that people vie to live in across the Home Counties,  paying that half and million for them, were built in the entire absence of detailed planning as under the Town and Country Planning Act 1947 and successors. That was also the last time the private sector produced the 300,000 units a year people say we need.

Time to abolish the Town and Country Planning Act 1947 and successors. The plan which will gain us the desired volume of housing that people actually want to live in, where they'd like to live. The solution, as so often, being that government does less. 

The Lancet says that zero booze is the healthiest lifestyle - Oh Aye?

The Lancet really surprises with the news that even a little sip of the Demon Booze is the Very Devil:

Overall, the study, which pooled data from 592 studies with 28 million participats, linked alcohol to almost 3 million deaths globally.

The research found it was the seventh leading risk factor for premature death and disease - and the leading cause of death before the age of 50. 

Researchers said the analysis found no safe level of alcoholic consumption, suggesting that going teetotal was the only way to avoid associated health risks. 

Given that this result is entirely different from everything anyone else has ever found out about booze then well, why? Sure, science does indeed say that new evidence outweighs, disproves, old theories. But that evidence had better be pretty good - the larger the claim the better the evidence needs to be.

We have to say that we're not convinced here. And do note that this is tentative as yet, this critique being possibly subject to later revision.

Conclusion

Alcohol use is a leading risk factor for disease burden worldwide, accounting for nearly 10% of global deaths among populations aged 15–49 years, and poses dire ramifications for future population health in the absence of policy action today. The widely held view of the health benefits of alcohol needs revising, particularly as improved methods and analyses continue to show how much alcohol use contributes to global death and disability. Our results show that the safest level of drinking is none. This level is in conflict with most health guidelines, which espouse health benefits associated with consuming up to two drinks per day. Alcohol use contributes to health loss from many causes and exacts its toll across the lifespan, particularly among men. Policies that focus on reducing population-level consumption will be most effective in reducing the health loss from alcohol use.

That is startlingly different from everything else we think we know about the effects of alcohol. So, why?

What intrigues us is the chart of deaths for the US. Page 2085 (yes, really) here. It lists those causes of death which are - usually that is, whether rightly or wrongly - said to be caused by alcohol consumption. What it doesn't do is list all causes of death. Reading from this listing of the leading US causes of death we note that pneumonia and Alzheimer's aren't there, while they are leading causes of death.

As above, this is tentative. But what it looks like to us is that they've added up all the deaths and diseases which could be attributable to alcohol. OK. And those rise monotonically with alcohol consumption. OK. And they've not looked at all causes of death to see whether alcohol consumption reduces the death rate from any other causes. That being the original claim in the first place, that yes of course booze causes problems even as it also salves some others.

It's as if someone looks at the effects of exercise and notes the costs in twisted ankles and strained backs but not the benefits in strengthened hearts and lower weights.

Again, as we say, this critique of ours is subject to revision as those who know more than we do pile in. But we're deeply, deeply, unconvinced of the finding here.  Perhaps the worst of it is that we know very well that there's a political movement to insist that no one should drink at all. We've seen other "research" making similar howlers to try and bolster that case. Meaning that we don't think we can trust research which produces results so amenable to that political campaign.

The entire field is so polluted by policy based evidence making that we assume that this is such. A pity really, as The Lancet did do science at one time but that well has rather been polluted.

Chris Snowdon's take is here.

An interesting illustration of the problems with booze

We're all aware that government guidelines on how much alcohol we might drink have something severely wrong with them. The latest report excised teetotallers from the numbers in order to get rid of that inconvenient fact that less booze can be as dangerous as more. The sweet spot of consumption being, in terms of all risks, more than the maximum we're advised to neck.

It's also true that current weekly limits are rapidly approaching what used to be considered a decent lunch. This might be granting too much to the wowsers. At which point we get this, the very latest research:

Alcohol revenue would decline by two-fifths, or £13 billion, if all drinkers were to comply with the recommended consumption limits, according to a study.

The research found that about two-thirds of alcohol sales in England are to heavy drinkers.

Drinkers who consume more than the Government's low-risk guideline of 14 units a week make up 25% of the population but provide 68% of alcohol industry revenue, according to a paper published in the journal Addiction.

The 4% of the population whose drinking is considered harmful - more than 35 units a week for women and more than 50 for men - account for almost a quarter (23%) of revenue, analysis by researchers at the Institute of Alcohol Studies (IAS) and the University of Sheffield's Alcohol Research Group suggests.

The report said the findings "raise serious questions about the conflicts of interest inherent to voluntary schemes and self-regulation".

It's easy enough to see what the next demand is. If voluntary schemes don't - in the minds of the prodnoses - work then there must be statutory and regulatory ones. It is not farfetched to think of alcohol rations, cards to be presented when asking for a snifter. Actually, that's already been suggested.

As to the basic observation, yes, that's how life works. A large portion of internet usage comes from those who use the internet a lot. Great chunks of road mileage come from those who drive a lot. Significant parts of meat consumption come from those on Atkins diets. People who do a lot of a thing tend to be a goodly portion of the thing being done. We'd be surprised to find that the sexually active and promiscuous are having a decent portion of the new sexual partners going on, would we? 

However, the thing which really struck us. There we've got - accurately enough too - the amounts which are actually harmful drinking, those 35 and 50 units a week. That is the sort of level, perhaps a little beyond, at which the risks significantly outweigh those of none at all.

So why are the government guidelines 14 units? The answer being that those limits are entirely invented, there's no medical justification at all. Just plucked from the air they are. Or from the prejudices of the anti-booze movement for all should know that alcohol is the very devil. 

Now for the important point. Yes, there's an education function to government, it's fair enough that we be told of scientific results. Not drinking from where the latrine flows out is good advice, it's saved many hundreds of millions of lives and it wasn't known before we were all told about it. It's not well enough known in some parts of the world today. But that is where that function ends, informing.

A free society allows properly informed people to do as they wish simply because that's what freedom is, it's the definition of liberty. People desire to drink more booze than is good for them physically? Well? 

Aid those who wish to drink less but find they cannot, most certainly, but our livers, ours to pickle if that's that we wish.

What are, quite frankly, lies about safe drinking annoy but what should enrage is this idea that we should be forced, forced, to consume as they think we should. We do actually see signs, protestations, saying "Our Bodies, Ourselves," books with titles like "Our Bodies, Our Souls." Booze is one of those issues over which this is all true.

Madsen Moment – Conservatism

The soul of parties that use the moniker conservative across the Western world appears to be up for grabs. Dr Pirie tackles just what Conservatism really means in this week's Madsen Moment.