Matt Ridley's Blog, page 38
August 24, 2014
Try free enterprise in Europe
My recent Times column was on the stagnation of European
economic growth rates:
The financial crisis was supposed to have
discredited the “Anglo-Saxon” model of economic management as
surely as the fall of the Berlin wall discredited communism. Yet
last week’s numbers on economic growth show emphatically the
opposite. The British economy is up 3.2 per cent in a year, having
generated an astonishing 820,000 jobs. We are behaving more like
Canada, Australia and America than Europe.
If you think one year is too short, consider that (as David Smith pointed out in the Sunday Times)
Britain’s GDP is now 30 per cent higher than it was in 1999,
whereas Germany, France and Italy are just 18 per cent, 17 per cent
and 3 per cent more prosperous respectively. For all Britain’s huge
debt burden, high taxes and chronic problems, we do still seem to
be able to grow the economy. Thank heavens we stayed out of the
euro.
The performance of the euro economies continues to be dismal. Last week’s news was that France is flatlining,
Germany shrinking slightly and Italy back in recession for its
third dip. Spanish unemployment is just a tad under 25 per cent.
The Greek economy continues to contract; even the Dutch economy is
bumping in and out of growth. The eurozone as
a whole is flat and teeters on the brink of debilitating
deflation. This at a time when the world economy, driven by Asia
and Africa, is roaring ahead at a forecast 3.7 per cent this year,
according to the International Monetary Fund.
The euro is not primarily to blame for eurosclerosis; there has
been plenty wrong with domestic policies in the individual
countries, though joining the euro allowed them to conceal their
mistakes for a while. But all the lessons point in the same
direction: public spending and dirigisme to stimulate growth does
not work, while limiting taxes and regulation to unleash growth
does.
Patrick Minford and Jiang Wang produced clear evidence a few years ago that “the
surest way to increase economic growth is to reduce government
spending and taxation”: as figures from the Organisation for
Economic Co-operation and Development confirm, a 10 per cent
increase in public spending produces a 0.5-1 per cent decrease in
growth rates. The encouragement of free enterprise is what has
always brought growth, from ancient Phoenicia to modern Mauritius,
from Renaissance Italy to Silicon Valley.
Poland’s economy has doubled in size since the fall of
communism, while Ukraine’s has stagnated, because Poland made a far
more urgent dash in the direction of free markets in labour,
capital and trade. Estonia has been the top performing of the
former Soviet colonies because Mart Laar, the historian who became
prime minister in 1992 at the age of 32, had read only one book on
economics, Milton Friedman’s Free to Choose and
was in his own words “so ignorant” that he thought
flat taxes, privatisation and the abolition of tariffs and
subsidies constituted normal policy in the West.
Mr Laar ignored the warnings from most Estonian economists, who
told him what he proposed was as “impossible as walking on water”.
There’s a common theme here. Germany’s postwar economic miracle
happened because Ludwig Erhard abolished rationing and freed up
markets in the teeth of expert advice. When the American general
Lucius Clay said his experts thought these policies were a bad
idea, Erhard replied “so do mine”, and did it anyway. When Sir John
Cowperthwaite turned Hong Kong into a low-tax, free-trade enclave
in the 1960s, he had to turn a blind eye to the instructions of his
LSE-educated masters in London. Indeed, he kept failing to send
them data so they could not see what was happening.
A recent analysis by three German economists for
the think-tank Politeia looked at the reasons for the economic
transformations of Ireland (from 1986), Sweden (1991), New Zealand
(1988), Chile (1974) and Brazil (1990), all of which resulted in
sustained bursts of rapid economic growth after long spells of
stagnation, and concluded that the causes in every case were
deregulation of goods and services markets, liberalisation of
labour markets, abolition of tariffs or subsidies, privatisation of
state enterprises and the encouragement of competition.
“Anglo-Saxon” stuff in every case.
Sweden is an interesting case, because many people still think
of it as showing an alternative route to prosperity than the
Anglo-Saxon one. Nima Sanandaji, a Kurdish-Swede, demonstrated the very opposite in a paper for
the Institute for Economic Affairs two years ago. He concluded:
“Sweden did not become wealthy through social democracy, big
government and a large welfare state. It developed economically by
adopting free-market policies in the late 19th century and early
20th century.”
Between 1870 and 1936, when it was a poster boy for Adam Smith,
Sweden had the fastest growth rate in the industrialised world and
spawned Volvo, Ikea, Ericsson, Tetra Pak and Alfa Laval. Then
between 1950 and 1990 it went from having an unusually small state
sector to having a very big one. The result was currency
devaluation, stagnation and slow growth, culminating in a
full-blown economic crisis in 1992 and a rapid fall down the
economic league tables. When it then cut taxes, privatised
education and liberalised private healthcare in the 1990s, it
rediscovered growth and sailed through the financial crisis in
pretty good shape.
Entire continents teach the same lesson. South America and now
Africa have both confirmed the hypothesis that state-directed
commerce leads to stagnation while free enterprise causes rapid
growth.
As the economic historian Deirdre McCloskey argues in her forthcoming
book Bourgeois Equality, the chief beneficiaries
of free enterprise revolutions are the poor. As a result of what
she calls “the great enrichment” since 1800, she says “millions
more have gas heating, cars, smallpox vaccinations, indoor
plumbing, cheap travel, rights for women, lower child mortality,
adequate nutrition, taller bodies, doubled life expectancy,
schooling for their kids, newspapers, a vote, a shot at university
and respect. Never had anything similar happened, not in the glory
of Greece or the grandeur of Rome, not in ancient Egypt or medieval
China.”
When Mahatma Gandhi was asked what he thought about western
civilisation, he said it would be a good idea. Likewise,
continental Europe’s approach to free enterprise should be to “give
it a try”. How many more years of self-imposed depression must the
European continent suffer before its political masters get over
their ideological prejudices against Anglo-Saxons and try reading a
little Adam Smith, Friedrich Hayek or Milton Friedman?
August 15, 2014
Reasons to be cheerful
The Times carried my article arguing that things are still going
well for the world as a whole even in a month of war, terror and
disease. I have illustrated it with two superb charts from ourworldindata.org, a website being developed
by the talented Max Roser.
Is this the most ghastly silly season ever? August 2014 has
brought rich pickings for doom-mongers. From Gaza to Liberia, from
Donetsk to Sinjar, the four horsemen of the Apocalypse — conquest,
war, famine and death — are thundering across the planet, leaving
havoc in their wake. And (to paraphrase Henry V), at their heels,
leashed in like hounds, debt, despair and hatred crouch for
employment. Is there any hope for humankind?
Consider the litany of horror that faces the world. A religious
war between militant Islam and its enemies is flaring all across
Eurasia, from Pakistan through Iraq, Syria, Palestine, Libya,
Somalia, South Sudan to Nigeria. In Ukraine a tinpot tyrant has
deliberately loosed a war of conquest and reconquest. In West
Africa a vicious pestilence spreads ever faster.
Think only of how often you have seen images of dead children
this summer: strewn across a cornfield in Ukraine, decapitated on a
street in Iraq, blown apart on a beach in Gaza, wounded in a
hospital in Syria, being buried in Liberia. The fate of the girls
kidnapped by Boko Haram in Nigeria is hardly any less horrible. Man
is a wolf to man.
In the world of money you can find plenty to cry about too.
Argentina has defaulted on its debt. Britain’s national debt has
doubled in four years. The Eurozone is in permanent recession and
teeters on the brink of its next crisis. Stock markets are
wobbling.
All true and all horrible. But the world is always full of
atrocity, violence, death and debt. Are things really worse this
year or are we journalists just reporting the clouds in every
silver lining? Remember the media does not give a fair summary of
what happens in the world. It tells you disproportionately about
the things that go badly wrong. If it bleeds, it leads, as they say
in newspapers. Good news is no news.
So let’s tot up instead what is going, and could go, right.
Actually it is a pretty long list, just not a very newsworthy one.
Compared with any time in the past half century, the world as a
whole is today wealthier, healthier, happier, cleverer, cleaner,
kinder, freer, safer, more peaceful and more equal.
The average person on the planet earns roughly three times as
much as he or she did 50 years ago, corrected for inflation. If
anything, this understates the improvement in living standards
because it fails to take into account many of the incredible
improvements in the things you can buy with that money. However
rich you were in 1964 you had no computer, no mobile phone, no
budget airline, no Prozac, no search engine, no gluten-free food.
The world economy is still growing every year at a furious lick —
faster than Britain grew during the industrial revolution.
Here's Max Roser's chart of the decline in the price of light
over two centuries:
The average person lives about a third longer than 50 years ago
and buries two thirds fewer of his or her children (and child
mortality is the greatest measure of misery I can think of). The
amount of food available per head has gone up steadily on every
continent, despite a doubling of the population. Famine is now very
rare. The death rate from malaria is down by nearly 30 per cent
since the start of the century. HIV-related deaths are falling.
Polio, measles, yellow fever, diphtheria, cholera, typhoid, typhus
— they killed our ancestors in droves, but they are now rare
diseases.
We tell ourselves we are miserable, but it is not true. In the
1970s there was a study that claimed to find that people grew less
happy as they got richer, but it was based on faulty data. We now
know that on the whole people are more satisfied with life as they
get wealthier, a correlation that holds between countries, within
countries and within lifetimes. Anyway, it’s better to be well fed,
healthy and unhappy than hungry, sick and unhappy. Here's Roser's
chart of happiness data:
Education
is in a mess and everybody’s cross about it, but consider: far more
people go to school and stay there longer than they did 50 years
ago. Besides, through a mysterious phenomenon called the Flynn
effect, IQ scores keep going up everywhere, especially in those
topics that have least to do with education, probably thanks to
better food, richer upbringing and so forth.
The air is much cleaner than when I was young, with smog largely
banished from our cities. Rivers are cleaner and teem with otters
and kingfishers. The sea is still polluted and messed with in every
part of the world, but there are far more whales than there were 50
years ago. Forest cover is increasing in many countries and the
pressure on land to grow food has begun to ease.
We think we are getting ever more selfish, but it is not true.
We give more of our earnings to charity than our grandparents did.
Violent crimes of almost all kinds are on the decline — murder,
rape, theft, domestic violence. So are capital and corporal
punishment and animal cruelty. We are less prejudiced about gender,
homosexuality and race. Paedophilia is no more prevalent, just
hushed up less.
Despite all the illiberal things our governments still try to do
to us, freedom is on the march. When I was young only a few
countries were democracies; the rest were run by communist or
fascist despots. Today there’s only a handful of the creeps left —
they could all meet in a pub: fat Kim, Castro the brother, Mugabe,
a couple of central Asians, the blokes from Venezuela and Bolivia,
the Belorussian geezer. Putin’s applying for membership. The
Chinese one no longer shows up.
The weather is not getting worse. Despite what you may have
read, there is no global increase in floods, cyclones, tornadoes,
blizzards and wild fires — and there has been a decline in the
severity of droughts. If you got the opposite impression, it’s
purely because of the reporting of natural disasters, which has
become a lot more hysterical. Besides, thanks to better
infrastructure, communications and technology, there has been a
steep decline in deaths due to extreme weather.
Globally, your probability of dying as a result of a drought,
flood or storm is 98 per cent lower than it was in the 1920s. As
Steven Pinker documented in his book The Better Angels
of Our Nature, the number of deaths in warfare is also
falling, though far more erratically. The ten years 2000-10 was the
decade with the smallest number of deaths in warfare since records
began in the 1940s. That may not last — indeed, it is looking like
this decade may be worse. But it may be better.
Here's Goklany's data on global deaths from extreme weather:
As for inequality, the world as a whole is getting rapidly more
equal in income, because people in poor countries are getting
richer at a more rapid pace than people in rich countries. That has
now been true for two decades, but it has accelerated since the
great recession. The GDP per capita of Mozambique is 60 per cent
higher than it was in 2008; that of Italy is 6 per cent lower. A
country like Mozambique has been out of the headlines recently and
now you know why: things are mostly going right there.
Writing my book The Rational Optimist in the
middle of a great recession that seemed to be bringing the world
economy to its knees was brave to the point of foolhardiness. But
if anything I was too cautious. The world bounced back from that
recession far faster than I expected and the pace of innovation and
improvement redoubled.
Britain, too, did better than I feared. We are growing faster
than any other major economy, we have seen the unemployment rate
defy even the most cheery forecasts in its rate of fall and we have
kept the country safer from terrorism than was true for most of my
life. Technologies that seem indistinguishable from magic keep
falling cheaply into our hands.
Of course, like anybody I can still talk myself into gloom.
Scotland could break away. Militant Islam could tear our
communities apart. European bureaucrats could strangle innovation
even more than they do already. When asked what I most worry about,
I always reply “bureaucracy and superstition” because these are
what brought down previous civilisations in Ming China or Abbasid
Arabia.
Be warned that being cheerful guarantees you will never be taken
seriously. The philosopher John Stuart Mill said: “Not the man who
hopes when others despair, but the man who despairs when others
hope, is admired by a large class of persons as a sage.”
August 14, 2014
Gamekeepers are conservationists
My column in the Times on 11th August:
Tomorrow sees the start of the red grouse shooting
season, a sport under attack as never before, with a petition to
ban it, and campaigns to get supermarkets to stop selling grouse
meat.
As somebody who lives in the rural north and knows the issue at
first hand, I am in no doubt that the opponents of grouse shooting
have it backwards. On both economic and ecological grounds, the
shooting of grouse is the best conservation practice for the
heathery hills of Britain. If it were to cease, most
conservationists agree that not only would curlews, lapwings and
golden plover become much scarcer, even locally extinct, but much
heather moorland would be lost to forest, bracken, overgrazing or
wind farms.
Be in no doubt: management for grouse is conservation. The
owners spend money to maintain the heather moors that constitute an
ecosystem found almost nowhere other than Britain. They prevent
overgrazing, re-establish heather, remove plantations of non-native
sitka spruce, eradicate bracken, manage drainage, periodically burn
long heather, kill foxes and crows, refuse to build subsidised wind
farms, and thus maintain the great open spaces of the Pennines and
parts of Scotland where people are free to walk. In the past decade
alone, moorland owners have regenerated 57,000 acres of
heather.
More than £50 million is spent on conservation by grouse moor
owners every year. That’s roughly twice as much as the Royal
Society for the Protection of Birds devotes to its entire
conservation efforts. There is no way the taxpayer would or should
stump up that kind of cash to look after heather moors. But
somebody has to: there is no such thing as a natural ecosystem in
this country and conservation requires human intervention.
Grouse moor owners recoup some of their costs by leasing
shooting to wealthy clients, who often fly in from abroad, fill the
local hotels and create crucial local employment. In the economy of
many Pennine dales, grouse shooting is irreplaceable, adding more
than £15 million a year nationally and supporting 1,500 full-time
jobs. It redistributes money from hedge-fund managers in the south
and overseas to some of the poorest parts of rural Britain. Much as
you might wish them to, rich folk won’t spend lots of money in the
Pennines to watch rare birds; but they will to shoot grouse.
Astoundingly, golden plover, curlews and lapwings, the three
most iconic wading birds of the uplands, live at five times the
density and have more than three times the breeding success on
moors with gamekeepers compared with moors without gamekeepers.
That this is because of gamekeeping was confirmed in a series of experiments by the
Game and Wildlife Conservation Trust near Otterburn in which
matching areas of moor were either keepered or not, then swapped
around after four years.
These birds would be at risk of dying out if it were not for
gamekeepers, as would black grouse, ring ouzels and merlins.
Nesting on or near the ground, such birds are vulnerable to foxes
and crows that take their young. With unnaturally high numbers of
foxes and crows in Britain — because of human roadkill and garbage
— the only way the birds can thrive is if somebody controls the
numbers of crows and foxes. The RSPB knows this and kills both
species on some of its reserves.
As a result, grouse moors in spring are alive with the calls of
birds, whereas the moors that are not managed for grouse are
ornithological deserts. In Wales, for example, lots of conservation
bodies try to manage the hills for birds, but curlews and golden
plover are very scarce, black grouse non-existent — in sharp
contrast to the grouse-rich Pennines. One grouse moor owner I spoke
to last week said he was happy to challenge the RSPB to an
ornithological audit by a neutral body of its upland reserves
versus his grouse moor.
The RSPB argues that the hen harrier, a hawk that preys on
grouse and breeds on moors, is under threat of extinction, because
gamekeepers persecute it. Yesterday saw a damp day of protest on
its behalf. In fact the British hen harrier population is stable at
about 630 pairs and is much higher than it was 100 years ago when
these birds were confined mainly to islands like the Orkneys.
Most of them are in Scotland. The only three successful pairs in
England this year were on or next to managed grouse moors. They are
not breeding on the RSPB’s English reserves because they too are
vulnerable to fox predation, so they need gamekeepers as much as
curlews do. On a Pennine grouse moor there is ample food — grouse
and other birds. On a Welsh bird reserve there’s just the odd
meadow pipit to eat. Because hen harriers breed in colonies, as a
1990s experiment at Langholm in Scotland found, they can quickly
build up (to 20 pairs in that case) and destroy the economy and
jobs on the grouse moor. The harriers themselves would then
collapse in numbers for lack of food. By the end of the experiment,
hen harriers at Langholm were back to two pairs.
You can see why gamekeepers dislike the idea of being done out
of a job by a bird that cannot thrive without their protection;
little wonder that some must occasionally be tempted to deter or
even kill harriers. A sensible compromise is on the table, and moor
owners are ready to sign up to it: they would allow low densities
of harriers on grouse moors, removing the excess chicks to
repopulate Wales or Cornwall, and providing “diversionary feeding”.
Everybody gains. All that’s needed is the RSPB’s agreement, but it
is being obdurate and demanding unworkable preconditions.
The red grouse, the bird at the heart of all this, is an amazing
creature. It’s wholly dependent on grazing heather, it cannot
survive in captivity, it lures people to invest heavily in
conservation in the north, which supports the economy and benefits
other wildlife, and it’s found nowhere else in the world — unlike
the hen harrier, which is common across two continents. The grouse
population can be heavily cropped, just like sheep, to provide
fine, free-range meat.
The campaign against grouse shooting makes no ecological or
economic sense. Surely it is not a cynical attempt to raid urban
wallets with an emotive anti-rich campaign like the RSPCA’s
campaign against foxhunting? Surely not.
-----
Post script: The data on the effect of gamekeepers on the
breeding success of round nesting birds are truly striking. In the
charts below, the red bars are with gamekeeping, the blue bars
without. (Source: here)
This chart shows the change in abundance as a result of the
introduction of gamekeeping:
August 12, 2014
Reasons to be fearful about Ebola
My Times column on Ebola:
As you may know by now, I am a serial debunker of
alarm and it usually serves me in good stead. On the threat posed
by diseases, I’ve been resolutely sceptical of exaggerated scares
about bird flu and I once won a bet that mad cow disease would
never claim more than 100 human lives a year when some “experts”
were forecasting tens of thousands (it peaked at 28 in 2000). I’ve
drawn attention to the steadily falling mortality from malaria and
Aids.
Well, this time, about ebola, I am worried. Not for Britain,
Europe or America or any other developed country and not for the
human race as a whole. This is not about us in rich countries, and
there remains little doubt that this country can achieve the
necessary isolation and hygiene to control any cases that get here
by air before they infect more than a handful of other people — at
the very worst. No, it is the situation in Liberia, Sierra Leone
and Guinea that is scary. There it could get much worse before
it
gets better.
This is the first time ebola has got going in cities. It is the
first time it is happening in areas with “fluid population
movements over porous borders” in the words of Margaret Chan, the
World Health Organisation’s director-general, speaking last Friday.
It is the first time it has spread by air travel. It is the first
time it has reached the sort of critical mass that makes tracing
its victims’ contacts difficult.
One of ebola’s most dangerous features is that kills so many
health workers. Because it requires direct contact with the bodily
fluids of patients, and because patients are violently ill, nurses
and doctors are especially at risk. The current epidemic has
already claimed the lives of 60 healthcare workers, including those
of two prominent doctors, Samuel Brisbane in Liberia and Sheik Umar
Khan in Sierra Leone. The courage of medics in these circumstances,
working in stifling protective gear, is humbling.
Inevitably, some health workers are fleeing the affected areas
and inevitably many families of victims are coming to see the
isolation wards as places of death to which they do not want their
loved ones taken. It does not help that doctors and hospitals are
now so associated with the disease that machete-wielding villagers
in Guinea have been refusing to allow doctors to enter some areas,
on the suspicion that they were bringing the disease.
So no wonder Dr Chan says the outbreak “is moving faster than
our efforts to control it. If the situation continues to
deteriorate, the consequences can be catastrophic in terms of lost
lives but also severe socio-economic disruption.” There is little
doubt that the ebola epidemic will have huge indirect effects,
through interrupting treatment and prevention for other serious
diseases, as well as through the dislocation of the economy of west
Africa.
Consider just one case, that of the woman who probably
first brought the virus to Liberia in March when she returned from
Guinea feeling unwell. She was cared for by her sister till she
died. The sister felt ill and took a communal taxi to Liberia’s
capital Monrovia on the way see her husband, which resulted in the
deaths of five other passengers in the taxi. She rode pillion on a
motorbike some of the way and the driver has not been traced. That
sort of thing is happening all the time.
I still maintain that ebola is very unlikely to cause a global
pandemic. As a disease of human beings it is too quick, too
virulent, too easy to contain — for its own good. With reasonable
precautions like hygiene and isolation, strictly enforced, it
fizzles out fast. This is true, not just of ebola, but of all the
haemorrhagic fevers, like the lassa, hanta and marburg viruses.
These have caught the imagination of scriptwriters because the
deaths they cause are so gory and the prognosis of those infected
so dire. However, they have never managed to create a pandemic —
unless the theory is right that the plague recorded by Thucydides
in 430BC, which supposedly came down the Nile from Africa, was
ebola. Lassa (from rodents) and marburg (from bats) flare up from
time to time in Africa, and hanta (also rodents) killed 121
soldiers during the Korean war.
The first and (until this time) worst recorded outbreak of
ebola, in Yambuku in Congo in 1976, was exacerbated by well-meaning
nuns running a remote clinic. They re-used needles to give quinine
injections to people with malarial symptoms and the early symptoms
of ebola are like malaria. Three quarters of those who died caught
the virus this way; four of the nuns also died. Today, the chances
of health workers making the problem worse are remote.
The more febrile kind of science writer is given to suggesting
that ebola is a sort of revenge from the ravished rainforest for
the destruction we have wrought on it. That is nonsense. Blood
samples from pygmies suggest that ebola outbreaks have been
happening sporadically for a very long time and killing apes as
well as people. If anything, it is intact forests, full of fruit
for bats to feed on, that represent the greatest reservoir. Bats
carry and reproduce the ebola virus very effectively, but are much
less affected by it and they are almost certainly its natural
host.
We do need to treat bats with caution. They have already given
us rabies, marburg virus and a morbillivirus in Australia that is
lethal to horses. Ebola is their deadliest gift. Given that a
quarter of all mammal species are bats, that they often share our
living spaces and they live like us in dense colonies, the chances
are they have more viruses to pass on. In the 1990s, a woman in an
animal sanctuary in Australia died from a bat-borne lyssavirus. (I
would forbid zoos and animal sanctuaries from handling bats in
tropical regions.)
Liberia and Sierra Leone are two of only six countries in the
world whose average per capita income is lower today than it was 50
years ago, and that is why they are so vulnerable to this epidemic.
The key lesson is not to slow or reverse development in rural
Africa. Quite the opposite. The sooner we can engage more of the
citizens of Liberia, Guinea and Sierra Leone in the global economy,
so they can get jobs in urban areas, afford decent healthcare and
begin to eat fast food rather than bushmeat, the better.
August 4, 2014
The coup d'etat of 1714 - when the Whigs won
I have a piece in the latest Spectator on the
tercentenary of King George I:
The centenary of the start of the first world war is getting
much more attention than the tricentenary of the accession of
George I, which also falls this week. As far as I can tell, no new
biographies of the first Hanoverian king are imminent, whereas
books on the great war are pouring forth. You can see why. The
replacement of a plump, if benign, queen by an ‘obstinate and
humdrum German martinet with dull brains and coarse tastes’
(Winston Churchill’s words), who presided over a huge financial
scandal and died unlamented after a short reign, need hardly detain
us.
But forget the royals and focus on what we might call the
reshuffle among politicians that accompanied the change. Here’s how
Henry St John, 1st Viscount Bolingbroke, described the last week of
July 1714 in a letter to Dean Swift: ‘The Earl of Oxford was
removed on Tuesday. The Queen died on Sunday. What a world this is,
and how does fortune banter us.’
The fall of the Jacobite-leaning Tories, led by Bolingbroke and
his rival and former friend Oxford, with a coup
d’état in the Privy Council by the Hanoverian-favouring Whigs,
led by the Duke of Shrewsbury, on 30 July turned out to be a key
moment in British history. It was never reversed, despite several
attempts. In its own way it was as significant as 1215 and
1688.
The Tory Bolingbroke, a dazzling orator and spectacular
libertine, had been stuffing positions of power with fellow
Jacobites since becoming secretary of state and overshadowing his
erstwhile ally the Earl of Oxford. But at an emergency privy
council meeting on 30 July following the Queen’s stroke, he found
himself outwitted by Shrewsbury, who unexpectedly summoned two
fellow Whigs, the Dukes of Argyll and Somerset. The council got the
barely conscious Queen to make Shrewsbury Lord Treasurer, then sat
late into the night dispatching messages to alert garrisons and
ensure that the Hanoverian succession was proclaimed.
Had Bolingbroke prevailed at that meeting, we would probably
have had a King James III, though there would almost certainly have
been a civil war (instead of the minor fiasco of the Fifteen).
Britain might have been more absolutist, more French influenced,
more Catholic-tolerant and less commercial. The stirrings of steam
in the north that were to start the industrial revolution — the
first faltering steps to turning heat into work — might have
fizzled. The Act of Union with Scotland, agreed to some years
earlier as part of the English insistence on the Hanoverian
succession, might have unravelled.
At least, so goes conventional wisdom. In Churchill’s words, the
outcome of that long meeting of the privy council was ‘No popery,
no disputed succession, no French bayonets, no civil war’.
However, there is another possibility. When not bonking,
Bolingbroke was a philosopher, a religious free thinker greatly
admired by Voltaire and Alexander Pope. His speeches and writings
were read with avidity by the American founding fathers, who
credited Bolingbroke with the idea that liberty means being free,
‘not of the law but by the law’. He invented the concept of an
official political opposition and saw it as his duty to prevent the
Whigs turning into a perpetual oligarchy. He proposed free trade
with France.
He was, in other words, a great deal more of an Enlightenment
figure than the Whig who replaced him and, thanks to the blind
support of George I and II, dominated politics for 20 years, while
filling his pockets with ill-gotten gains: Robert Walpole.
Thus the cartoon version of history in which Whigs and
Hanoverians brought liberty, parliament, Protestantism and trade,
while Tories and Stuarts would have brought absolutism, Popery and
civil war, may not be right. You cannot quite help wondering if a
Bolingbroke ascendancy might have given England a more vigorous
Enlightenment, too, to rival those in France and Scotland. It has
always puzzled me that the stars of the Enlightenment — Voltaire,
Diderot, Hume, Smith and co. — included plenty of Scots and French,
but no Englishmen.
Had Bolingbroke persuaded James Edward Stuart to turn
Protestant, as he had tried to, then many British people would have
welcomed a Stuart king. The idea of a German-speaking monarch was
not at all popular. Shrewsbury’s coup might well have failed.
As it was, it was a close-run thing. There were plenty of
Protestants who favoured James. I recently found out that my
ancestor, who was Tory mayor of Newcastle that year, refused to
declare the accession of George despite being a staunch Protestant.
A rival faction did declare it, so Richard Ridley sent his thugs to
stamp it out, resulting in a Friday night riot on the Quayside
(nothing much has changed).
Still, it all worked out in the end. Britain may not have loved
its new king, nor the corrupt grandees who ruled in his name and
promptly debauched the currency in the South Sea Bubble. But George
did give sanctuary to Voltaire when he was exiled from France, and
gradually the country did take advantage of the largest free-trade
area in Europe (England and Scotland) to sow the seeds of
prosperity and incubate freedom.
Bolingbroke’s most famous work, The Idea of a Patriot
King, was written at Alexander Pope’s behest much later in
1738 to influence George I’s grandson Frederick, Prince of Wales,
into being a monarch who rose above faction, was a father to his
country and championed trade.
Which, if you think about it, is roughly what we have now.
This article first
appeared in the print edition of The Spectator magazine,
dated 2 August
2014
August 2, 2014
Priorities and goals for aid
My recent essay in the Wall Street Journal
discusses how to prioritise development aid:
In September next year, the United Nations plans to choose a
list of development goals for the world to meet by the year 2030.
What aspirations should it set for this global campaign to improve
the lot of the poor, and how should it choose them?
In answering that question, U.N. Secretary-General Ban Ki-moon
and his advisers are confronted with a task that they often avoid:
setting priorities. It is no good saying that we would like peace
and prosperity to reach every corner of the world. And it is no
good listing hundreds of targets. Money for foreign aid, though
munificent, is limited. What are the things that matter most, and
what would be nice to achieve but matter less?
The origin of this quest for global priorities goes back to
2000, when Mr. Ban's predecessor, Kofi Annan, picked a set of
"Millennium Development Goals," eight challenges to be met by 2015,
which were adopted by world leaders. Although some of these goals
were woolly, the very brevity of the list and the deadline itself
meant that they really did catch the world's imagination and force
the aid industry to be more selective.
Most of the original Millennium Development Goals will have been
met or nearly so by 2015. Since 2000, for example, the number of
people living in extreme poverty and hunger around the world will
have been cut in half—an astonishing achievement. Other goals
included universal primary education, gender equality, reductions
in child mortality, improvements in maternal health, progress
against HIV and malaria, environmental sustainability and (most
vaguely) a "global partnership for development."
The lesson, surely, from this first round of setting development
goals is the need to be even more ruthlessly selective next time. A
list of eight goals is too long for most outsiders to remember.
When I asked several of my colleagues in the British Parliament,
they remembered only three to five. Several development experts I
spoke to say that the new list should have just five discrete,
quantitative, achievable goals.
Only Mr. Ban can make that happen, says Charles Kenny, a senior
fellow at the Center for Global Development in Washington, D.C.,
who observes that you should "never ask a committee to write
poetry." Mr. Kenny told me: "There is one person who can bring the
poetry. The U.N. secretary-general has to edit down with an ax, not
a scalpel. Without strong intervention from Ban Ki-moon, there is
extremely limited prospect for simplification."
Goal: Boost preprimary education,
which costs little and has lifelong benefits by getting children
started on learning. European Pressphoto Agency
So far, however, the process of deciding on the 2030 goals is
short on poetry. There is not just one committee on the job but
several—the most prominent of which is called the Open Working
Group, or OWG, which has already been meeting off and on for more
than two years. The OWG "stream"—and keep in mind that other U.N.
groups are also producing streams of their own—has so far managed
to whittle its list of possible targets down to 169. It is an
absurdly long list, and each time the results of its deliberations
are published, every pressure group checks to make sure its
favorite goal is still in there and makes a fuss if it is not.
What Mr. Ban needs is an objective way of paring down the list.
In doing so, I would recommend to him an unlikely ally: Bjorn
Lomborg, a T-shirt-wearing, vegetarian, Danish political scientist
who shot to fame in 2001 with a book called "The Skeptical
Environmentalist," which infuriated those who support environmental
protection at all costs, including the welfare of the poor.
Mr. Lomborg is the founder of an international think tank called
the Copenhagen Consensus Center. He has invented a useful method
for dispassionately but expertly deciding how to spend limited
funds on different priorities. Every four years since 2004, he has
assembled a group of leading economists to assess the best way to
spend money on global development. On the most recent occasion, in
2012, the group—which included four Nobel laureates—debated 40
proposals for how best to spend aid money.
The goal was simple: to create a cost-benefit analysis for each
policy and to rank them by their likely effectiveness. For every
dollar spent, how much good would be done in the world?
The Copenhagen Consensus Center process has won world-wide
respect for its scrupulously fair methods and startling
conclusions. Its 2012 report, published in book form as "How to
Spend $75 Billion to Make the World a Better Place," came to the
conclusion that the top five priorities should be nutritional
supplements to combat malnutrition, expanded immunization for
children, and redoubled efforts against malaria, intestinal worms
and tuberculosis.
Their point wasn't that these are the world's biggest problems,
but that these are the problems for which each dollar spent on aid
generates the most benefit. Enabling a sick child to regain her
health and contribute to the world economy is in the child's
interest—and the world's.
The numbers produced by this exercise are eye-catching. Every
dollar spent to alleviate malnutrition can do $59 of good; on
malaria, $35; on HIV, $11. As for fashionable goals such as
programs intended to limit global warming to less than two degrees
Celsius in the foreseeable future: just 2 cents of benefit for each
dollar spent.
Nor is this just about the cold tabulation of dollars and cents.
The calculus used by the Copenhagen Consensus also includes such
benefits as avoided deaths and sickness and potential environmental
benefits, including forestalling climate change.
The Copenhagen experts use strips of paper on which are written
different priorities along with cost-benefit ratios, and they are
invited to move them up and down as they debate the academic
evidence. In setting priorities, they also take into account the
feasibility of scaling up interventions and the risk of
corruption.
Of course, when the U.N. is contemplating its choices for the
next set of global development goals, cost-benefit isn't the only
criterion. In South Africa, for instance, HIV is a much bigger
problem than malaria, so different regions will have different
concerns. But ranking the interventions does concentrate the
mind.
Surprising as it may seem, the global-aid industry has rarely
done such cost-benefit analysis. People in this line of work
generally recoil from such rankings as a heartless exercise
implying discrimination against still-worthy global goals. The aid
industry often seems implicitly to take the view that funds are
unlimited and that spending on one priority doesn't crowd out
spending on another. But this is patently not the case: The
problems are far bigger than the available budget and will remain
so even if the world's rich countries ever meet their 35-year-old
goal of spending 0.7% of their GNP on development aid.
In December last year, Mr. Lomborg came to New York to address
the U.N. Open Working Group's ambassadors directly. He handed them
his strips of paper and asked them to put them down in preferred
order. It was an eye-opening exercise in a place where people are
accustomed to saying, in diplomatic earnest, "Everything is
important."
Then, over eight days in June, Mr. Lomborg got a group of 60
leading economists to work through all the OWG's putative
development targets for 2030 (there were more than 200 of them at
the time), making a quick assessment of which were good value for
money. The result, now available online, is a
document that assigns a color code to each target: green
(phenomenal value for money), pale green (good), yellow (fair),
gray (not enough known) and red (poor).
[Here are some of the Copenhagen
Consensus Center group's rankings.]
At the conclusion of this process, the group had 27 "phenomenal"
green values and 23 "poor" red values, with all the rest in
between.
Champions of aid aren't used to having their homework marked in
this stark fashion, and some didn't like it at first. As Ambassador
Elizabeth M. Cousens, the U.S. representative to the U.N. Economic
and Social Council, told Mr. Lomborg, "I really don't like you
putting one of my favorite targets in red." But she added, "I'm
glad you're saying it, because we all need to hear economic
evidence that challenges us."
Having gone through this useful document myself, I found myself
in full sympathy with those forced to choose among them. But at
least this sort of analysis provides some rigor and direction.
What would my own list of five 2030 goals look like, based on
the work of the Copenhagen Consensus group?
1. Reduce malnutrition. When children get better food, they
develop their brains, stay in school longer and end up becoming far
more productive members of society. Every dollar spent to alleviate
malnutrition brings $59 of benefits.
2. Tackle malaria and tuberculosis. These two diseases
debilitate huge populations in poor countries, but they are largely
preventable and curable. In the most harshly affected countries,
two people often do one person's work because one of them is sick.
Benefit to cost ratio: 35 to 1.
3. Boost preprimary education, which costs little and has
lifelong benefits by getting children started on learning. 30 to
1.
4. Provide universal access to sexual and reproductive health,
which would save the lives of mothers and infants while enabling
women to be more economically productive. It would also lower
birthrates (when fewer children die, people have fewer children).
Benefits could be as high as 150.
5. Expand free trade. This isn't considered sexy in the
development industry, and it may seem remote from humanitarian
issues, but free trade often delivers phenomenal improvements to
the welfare of the poor in surprisingly quick time, as the example
of China has demonstrated in recent years. One of the discoveries
of the Copenhagen Consensus process is that incremental goals such
as expanding free trade are often better than supposedly
"transformational" goals. A successful Doha Round of the World
Trade Organization could deliver annual benefits of $3 trillion for
the developing world by 2020, rising to $100 trillion by the end of
the century.
The development goals of least value, according to the
Copenhagen process, include the self-contradictory call for higher
agricultural productivity with less environmental impact. Other bad
investments are less obvious but would actually hurt the poor. For
example, equal access to affordable tertiary education may sound
good in principle, but in many developing countries, it amounts to
a policy of having the mass of poor people pay for the college
education of the rich. Other goals—such as "sustainable
tourism"—are simply too narrow and ill-defined to merit
consideration on a list of urgent priorities.
One much-favored goal in the list generated by the U.N.'s Open
Working Group comes out especially badly: the idea of providing
gender-disaggregated data to help women. Not only do we have much
of the data (and it is very costly to gather more), but how, say
the Copenhagen experts, would you define the gender-disaggregated
value of a cow owned by a family of five?
Those who fear that the rankings might reflect Mr. Lomborg's own
prejudices will be relieved. He convened the economists, to be
sure, but they are the ones who did the color coding.
Mr. Lomborg accepts the basic conclusions of today's climate
science, but he is known to be skeptical about many current
policies to avert climate change. Still, the experts he brought
together conclude that phasing out fossil-fuel subsidies is a
"phenomenal" value. They also find excellent value in programs
meant to develop resilience and adaptive capacity in response to
climate-induced hazards.
But they judge it poor value, for the world's poor, to attempt
either to double the share of renewable energy in the global energy
mix or to hold the increase in global average temperature below a
certain level in accordance with international agreements. This is
because the experts think that allowing emissions to rise initially
while investing in rapid advances in energy technology is a much
better idea than trying to limit emissions now with today's
expensive renewables.
Indeed, one of the world's most pressing health problems, and
the one most conspicuously missing from Mr. Annan's original
development goals in 2000, is the annual death toll of more than
four million people due to indoor air pollution. This enormous,
abiding problem is attributable to the fact that so many of the
world's poor lack access to affordable (that is,
fossil-fuel-generated) electricity and therefore cook over burning
wood or dung.
This most recent exercise by the Copenhagen Consensus was, Mr.
Lomborg admits, "quick and dirty," intended to catch the attention
of the Open Working Group before it wraps up its work for the
summer. But in the coming months, Mr. Lomborg's group will publish
thousands of peer-reviewed pages, describing costs and benefits for
all the most important U.N. targets. With the help of three Nobel
Laureates, the group will produce a definitive report with ranked
priorities and deliver it to the U.N.
Figuring out the best way to help the world's poor isn't like
solving a math problem. There are not right and wrong answers. But
there are better and worse answers, and the only way to assign
those priorities is to set aside our sentimental commitments and do
the hard work of assessing costs and benefits.
July 31, 2014
Renewable energy is not working
My Times Column explores why renewable energy has
been so disappointing.
On Saturday my train was diverted by engineering
works near Doncaster. We trundled past some shiny new freight
wagons decorated with a slogan: “Drax — powering tomorrow: carrying
sustainable biomass for cost-effective renewable power”.
Serendipitously, I was at that moment reading a report by the chief scientist at the
Department of Energy and Climate Change on the burning of wood in
Yorkshire power stations such as Drax. And I was feeling
vindicated.
A year ago I wrote in these pages that it made no sense for
the consumer to subsidise the burning of American wood in place of
coal, since wood produces more carbon dioxide for each
kilowatt-hour of electricity. The forests being harvested would
take four to ten decades to regrow, and this is the precise period
over which we are supposed to expect dangerous global warming to
emerge. It makes no sense to steal beetles’ lunch, transport it
halfway round the world, burning diesel as you do so, and charge
hard-pressed consumers double the price for the power it
generates.
There was a howl of protest on the letters page from the chief
executive of Drax power station, which burns a million tonnes of
imported North American wood a year and plans to increase that to 7
million tonnes by 2016. But last week, Dr David MacKay’s report
vindicated me. If the wood comes from whole trees, as much of it
does, then the effect could be to increase carbon dioxide
emissions, he finds, even compared with coal. And that’s allowing
for the regrowth of forests.
Despite the best efforts of the Conservatives to rein in their
Lib Dem colleagues, the renewable-energy bandwagon careers onward,
costing ever more money and doing real environmental harm, while
producing trivial quantities of energy and risking blackouts next
winter. People keep telling me it’s no good being rude about all
renewables: some must be better than others. Well, I’m still
looking:
Tidal power remains a (literal) non-starter; if you ask
ministers why nothing has been built, they say it’s not for want of
proffering ludicrously generous subsidies on our behalf. Yet still
no takers.
Wave power: again, the sky’s the limit for what the government
will pay if you can figure out how to make dynamos and generators
survive the buffeting of waves, corrosion of salt and encrustation
of barnacles. Nothing doing.
Geothermal: perhaps great potential in the future for heating
homes through district heating schemes, though expensive here
compared with Iceland, but not much use for electricity. Air-source
and ground-source heat pumps, all the rage a few years ago, have
generally proved more costly and less effective than advertised,
but they are getting better. Trivial contribution so far.
Solar power: one day soon it will make a big impact in sunny
countries, and the price is falling fast, but generating for the
grid in cloudy Britain where most power is needed on dark winter
evenings will probably never make economic sense. Covering fields
in Devon with solar panels today is just ecological and economic
vandalism. Solar provides about a third of one per cent of world
energy.
Offshore wind: Britain is the world leader, meaning we are the
only ones foolish enough to pay the huge subsidies (treble the
going rate for electricity) to lure foreign companies into tackling
the challenge of erecting and maintaining 700ft metal towers in
stormy seas. The good news is that the budget for subsidising
offshore wind has almost run out. The bad news is that it is
already costing us billions a year and ruining coastal views.
Onshore wind: one of the cheapest renewables but still twice as
costly as gas or coal, it kills eagles and bats, harms tourism,
divides communities and takes up lots of space. The money goes from
the poor to the rich, and the carbon dioxide saving is tiny,
because of the low density of wind and the need to back it up with
diesel generators. These too now need subsidy because they cannot
run at full capacity.
Hydro: cheap, reliable and predictable, providing 6 per cent of
world energy, but with no possibility for significant expansion in
Britain. The current vogue for in-stream generation in lowland
streams in England will produce ridiculously little power while
messing up the migration of fish.
Anaerobic digestion: a lucrative way of subsidising farmers (yet
again) to grow perfectly good food for burning instead of eating.
Contrary to myth, nearly all the energy comes from crops such as
maize (once fermented into gas), not from food waste.
Expensive.
Waste incineration: a great idea. Yet we are currently paying
other countries to take it off our hands and burn it overseas. If
instead we burned it at home, we would make cheap, reliable
electricity. But Nimbys won’t let us.
Over the past ten years the world has invested more than $600 billion in wind power
and $700 billion in solar power. Yet the total contribution those
two technologies are now making to the world primary energy supply is still less than
2 per cent. Ouch.
If we had spent that sum on research, and steadily replaced coal
with gas as a source of electricity, we would have done far more to
cut carbon emissions and kept prices low. A new report by Charles Frank of the Brookings
Institution has come to the startling conclusion that if you encourage gas
to replace coal, you get fewer emissions per dollar spent than if
you use wind or solar.
In Mr Frank’s words: “Solar and wind facilities suffer from
a very high capacity cost per megawatt, very low capacity factors
and low reliability, which result in low avoided emissions and low
avoided energy cost per dollar invested.” In short, we are picking
losers.
I would not suggest Drax goes back to burning only coal, partly
because I have a vested interest in the coal industry and partly
because more than 40 per cent of the coal we burn in this country
comes from Russia, so we are more exposed to Vladimir Putin for our
coal than for our gas. The answer is staring us in the face. Gas is
the cheapest clean way of making electricity, and we are sitting on
one of the world’s richest shale-gas fields. Yet investment in
gas-fired power is deterred by the government’s preference for
renewables.
July 24, 2014
Atheists and Anglicans could unite against intolerance
My Times column is on religion in schools:
We now know from Peter Clarke’s report, published today but
leaked last week, that there was indeed “co-ordinated, deliberate
and sustained action to introduce an intolerant and aggressive
Islamist ethos into some schools” in Birmingham.
Whistleblowers first approached the British Humanist Association
in January with such allegations, weeks before the appearance of
the Trojan Horse letter. The BHA (of which I should declare I am a
“distinguished supporter” though I’ve never done much to deserve
this accolade) properly passed on the information to the Department
for Education.
Pavan Dhaliwal, of the BHA, has made the awkward point that much
of what went on in the Park View Trust schools would have been
permissible if the schools had been designated “faith schools”. The
BHA campaigns against the very existence of state-funded faith
schools, pointing out that Britain is one of only four countries in
the world to allow religious selection in admissions to
state-funded schools. The others are Estonia, Ireland and
Israel.
In short, we can hardly be shocked to find religious
indoctrination going on in some schools if we encourage segregation
on the basis of faith. Since 2000 the proportion of secondary
schools that are legally religious has increased by 20 per cent,
and their freedom of action has greatly increased. The best way to
prevent young girls in Birmingham being told that “if a woman said
no to sex with her husband then angels would punish her from dusk
till dawn”, as happened in Birmingham, is to leave religious
practice — though not education about religion — out of school
altogether.
I know such a view is considered intolerant, even bigoted — a
charge frequently levelled at non-believers. “The trouble with that
Richard Dawkins”, a lay preacher said to me some years ago, “is
that he’s welcome to his views, but I don’t like him forcing them
on others.” Passing up the temptation to point out his own
hypocrisy as a preacher, I gently reminded him that, whereas I had
to go to prayers or chapel every day at my school, nobody has ever
been forced to read Richard Dawkins on atheism.
August sees a
great global gathering of atheists and humanists in Oxford for
the World Humanist Congress, the first time this body has met in
Britain since 1978. Professor Dawkins will be on the stage, along
with a galaxy of infidel stars, including the Nobel prizewinner
Wole Soyinka, Philip Pullman, Jim al-Khalili, Nick Clegg and the
Bangladeshi blogger Asif Mohiddun, who was attacked and stabbed in
the back, shoulder and chest by a group of radical religious
fundamentalists because of his criticism of Islam.
Not there in person will be Mubarak Bala, the Nigerian detained on a
psychiatric ward for being an atheist, whose case has been
highlighted by the International Humanist Ethical Union. His father
had Mr Bala sectioned for expressing doubts about religion and he
got out, two weeks ago, only because of a strike at the hospital.
Nor will Alexander Aan— the scientist in Indonesia who
was arrested and imprisoned for two years for expressing doubts
about God — be present. But many similar activists from Africa and
Asia will be there, including Gululai Ismail, who runs the Aware
Girls project in northwest Pakistan, challenging patriarchy and
religious extremism, and under constant threat of violence. It was
her organisation that Malala Yousafzai was working for when shot by
the Taliban.
It is clear that the kind of rational scepticism that we British
have been tolerating for three centuries is resulting in terrible
persecution throughout the Muslim world, and it is getting worse. I
say we tolerate atheism here, and we do, but still grudgingly.
Atheists lose count of the number of times we are told we are
lacking in imagination and wonder, or that we just don’t see the
human need for spirituality, or that we must have trouble
justifying morality.
British Christians are generally prepared to be much ruder about
atheism than they are about Islam. Some of the stuff Professor
Dawkins has to read about himself would be condemned as hate speech
if said about a Muslim. This is partly because atheists do not
threaten our critics with violence, whereas any “Islamophobic”
remark or cartoon leads to death threats. It is also because
Christians are continually trying to make common cause with other
religions in defence of “faith” as a source of morality and harmony
in the world. Did I dream it, or did a recent archbishop muse about
the virtues of Sharia?
Anglicanism is a mild and attenuated form of the faith virus and
may even act as a vaccine against more virulent infections, but
Christianity is becoming more evangelical in response to its global
competition with Islam. This has always happened in religious
history: where religions compete, they become more extreme — the
crusades, the 30-years war, Ulster.
So for all the pious talk of “faith communities”, the two
religions are not on the same side. To combat the rise of radical
Islam and radical Christianity, we should try the secular,
free-thinking approach. Mild Anglicanism should make common cause
with humanists in defence of tolerance.
The experience of the past three centuries is that if lots of
people stop believing in gods, they do not become less moral. On
the contrary, the number of people attending church has gone down
at about the same rate as the number of people who commit violent
crimes. I am not suggesting a causal connection — though I suspect
religious people would if the trends were different — but these
facts give the lie to the idea that godlessness leads to
immorality. (And don’t tell me that communist regimes were
irreligious — they enforced a worship of their leaders with all the
techniques and fervour of religion.)
Unlike the almost triumphalist mood among atheists in the 1960s,
when Francis Crick foresaw the end of religion and started a
competition for what to do with the college chapels in Cambridge,
rationalists no longer expect to get rid of religion altogether by
explaining life and matter: they aim only to tame it instead, and
to protect children from it. Nonetheless, they are slowly winning:
witness the fact that more than 12 per cent of
funerals in this country are now humanist in some form. And
humanists are showing no signs of turning intolerant, let alone
violent.
July 17, 2014
On Slippery Slopes
My Times column tackles the misleading metaphor of the slippery
slope:
Who first thought up the metaphor of the slippery
slope? It’s a persistent meme, invoked in many a debate about
ethics, not least over the assisted dying bill for which I expect
to vote in the House of Lords on Friday. But in practice, ethical
slopes are not slippery; if anything they are sometimes too
sticky.
It is in genetics and reproduction that the slippery slope
argument gets used most relentlessly. The latest science to suffer
from the slippery-slope metaphor is mitochondrial replacement
therapy (MRT), a substantially British innovation, which promises
to cure human mitochondrial diseases by replacing the mother’s cell
batteries with those from a donor. It will be up to regulators to
decide if the technique is safe, but first it is up to parliament
to decide if it is ethically acceptable.
The only real objection to this intervention to allow certain
people to have their own children without horrible diseases is that
it is, in the words of one critic, “a slippery slope to
human germline modification”. Change the 0.002 per cent of genes
that are in the mitochondrion and what’s to stop somebody one day
changing the 99.998 per cent of genes in the nucleus of the
egg?
Here’s a perennial opponent of genetics, David
King, of Human Genetics Alert, on MRT: “Once we’ve crossed this
crucial ethical line, which says that we shouldn’t create babies
that have been genetically altered, it becomes very difficult to
then stop when the next step is wanted and then the next step after
that and we will eventually get to this future that everyone wants
to avoid of designer babies.”
But why would it be very difficult to stop? If MRT is legalised,
it would still be illegal to use it for any purpose other than
prevention of mitochondrial genetic disease.
A real slippery slope is a muddy hillside where one small,
apparently safe, step can lead to a slide to the bottom on your
backside. This is because, the physicists tell me, the static
co-efficient of friction is greater than the kinetic co-efficient
of friction, so it takes more force to start a bottom sliding than
to keep it going.
Look around current debates and it seems we stand atop veritable
mountain ranges of slippery slopes. Assisted dying will lead to
widespread euthanasia. Gay marriage will lead to approved
bestiality. Artificial life will lead to biological warfare. But
metaphors can mislead. There is no equivalent in the world of
politics to that change in the co-efficient of friction that
happens on muddy hillsides. It is easy to stop half way down a
moral slope; it’s hard not to. Each step meets fierce opposition
however well the previous step went.
When we do carry on down an ethical slope, progressively
changing the moral code, it’s not because it is slippery and we
wish we could stop, but because we have collectively decided we
want to go further at each stage. The Great Reform Bill was a step
on the road to universal adult suffrage, as many conservatives at
the time feared it would be, but it was hardly a slippery slope,
more of a lengthy struggle. The legalisation of homosexuality was
indeed a step on the road to gay marriage, as many religious people
feared at the time, but only because society chose to take each
subsequent step.
Over 40 years we have repeatedly been promised that bad things
will come of interfering in reproduction, but so far the good has
vastly outweighed the bad. The invention of in-vitro fertilisation
in the 1970s was much feared by many people as the precursor to
eugenics — people would use the technology to have superior babies
by using the sperm of celebrities. Wrong: the demand for
high-performance donor fathers is all but non-existent; the
technology is used almost entirely to help people have their own
babies, to cure the cruel disease of infertility.
Then in the 1980s research on embryos was going to lead
inexorably to the cheapening of life. It did the opposite, leading
to the development of techniques such as pre-implantation genetic
diagnosis (PGD), in which inherited diseases could be avoided,
reducing misery for thousands. But then PGD was going to lead to
eugenics for the rich, who would pick the best genes for their
babies. Not so. There is very little interest in using PGD to
“improve” normal genomes rather than avoid faulty ones.
In the 1990s, there was one technology that overreached. Gene
therapy — the attempt to replace a faulty gene in a particular
tissue using a virus to deliver a new version — did make an early
mistake, contributing to the deaths of some patients. But now the
technique is safely doing good on a growing scale.
Earlier this year, it was announced that six people with a
previously incurable form of blindness had improved their sight
after gene therapy. This follows successful gene-therapy trials on
children with various life-threatening inherited conditions.
After so many decades of seeing genetics and reproductive
technologies prove more beneficial and less open to abuse than
expected, you might think an intervention such as MRT had earned
the benefit of the doubt. Indeed, you might think that the
conservative side of this debate would have seen enough to change
its mind and recognise that the alleviation of suffering should
take priority over the pious recitation of moral platitudes and
appeals to faulty metaphors about muddy hillsides.
You might even, in a sense, think that the slope should be a bit
more slippery. If each step proves more beneficial and less harmful
to humanity than expected, then the next step should be taken
faster. But this is not the way such debates work. The track record
of previous innovations in medicine is ignored when we decide each
new possibility. The slope is sticky, not slippery.
It staggers me that the resistance to new techniques of genetic
and reproductive medicine comes mainly from the right, and the
religious right at that. Where, pray, in the Bible or Koran does it
say that it is better that a child — or an old person — should
suffer than that the sanctity of natural genomes and natural
ailments be interfered with? And all just in case some vague and
implausible crime be facilitated in the distant future. Slippery
slopes are red herrings.
July 9, 2014
The BBC and balance
My Times column on the BBC's unbalanced
environmental coverage:
The BBC’s behaviour grows ever more bizarre.
Committed by charter to balanced reporting, it has now decided
formally that it was wrong to allow balance in a debate between
rival guesses about the future. In rebuking itself for having had
the gall to interview Nigel Lawson on the Today programme about
climate change earlier this year, it issued a statement containing
this gem: “Lord Lawson’s views are not supported by the evidence
from computer modelling and scientific research.”
The evidence from computer modelling? The phrase is an oxymoron.
A model cannot, by definition, provide evidence: it can provide a
prediction to test against real evidence. In the debate in
question, Lord Lawson said two things: it was not possible to
attribute last winter’s heavy rain to climate change with any
certainty, and the global surface temperature has not warmed in the
past 15 to 17 years. He was right about both, as his debate
opponent, Sir Brian Hoskins, confirmed.
As for the models, here is what Dr Vicky Pope of the Met Office
said in 2007 about what their models predicted: “By 2014, we’re
predicting that we’ll be 0.3 degrees warmer than 2004. Now just to
put that into context, the warming over the past century and a half
has only been 0.7 degrees, globally . . . So 0.3 degrees, over the
next ten years, is pretty significant . . . These are very strong statements about what will happen over the
next ten years.”
In fact, global surface temperature, far from accelerating
upwards, has cooled slightly in the ten years since 2004 on most
measures. The Met Office model was out by a country mile. But the
BBC thinks that it was wrong even to allow somebody to challenge
the models, even somebody who has written a bestselling book on
climate policy, held one of the highest offices of state and
founded a think-tank devoted to climate change policy. The BBC
regrets even staging a live debate between him and somebody who
disagrees with him, in which he was robustly challenged by the
excellent Justin Webb (of these pages).
And why, pray, does the BBC think this? Because it had a
complaint from a man it coyly describes as a “low-energy
expert”,
Mr Chit Chong, who accused Lord Lawson of saying on the
programme that climate change was “all a conspiracy”.
Lawson said nothing of the kind, as a transcript shows. Mr
Chong’s own curriculum vitae boasts that
he “has been active in the Green party for 25 years and was the
first Green councillor to be elected in London”, and that he “has a
draught-proofing and insulation business in Dorset and also works
as an environmental consultant”.
So let’s recap. On the inaccurate word of an activist politician
with a vested financial and party interest, the BBC has decided
that henceforth nobody must be allowed to criticise predictions of
the future on which costly policies are based. No more appearances
for Ed Balls, then, because George Osborne’s models must go
unchallenged.
By the way, don’t bother to write and tell me that Lord Lawson
is not a scientist. The BBC also rebuked itself last week for
allowing an earth scientist with dissenting views on to Radio 4.
Professor Bob Carter was head of the department of earth sciences
at James Cook University in Australia for 17 years. He’s published
more than 100 papers mainly in the field of paleoclimatology. So
bang goes that theory.
The background to this is that the BBC recently spent five years
fighting a pensioner named Tony Newbery, including four days in
court with six lawyers, to prevent Mr Newbery seeing the list of 28
participants at a BBC seminar in 2006 of what it called “the best
scientific experts” on climate change.
This was the seminar that persuaded the BBC it should no longer
be balanced in its coverage of climate change. A blogger named
Maurizio Morabito then found the list on the internet anyway. Far
from consisting of the “best scientific experts” it included just
three scientists, the rest being green activists, with a smattering
of Dave Spart types from the church, the government and the
insurance industry. Following that debacle, the BBC
commissioned a report from a geneticist, Steve Jones, which it
revisited in a further report to the BBC Trust last week. The Jones
report justified a policy of banning sceptics under the term “false
balance”. This takes the entirely sensible proposition that
reporters do not have to, say, interview a member of the Flat Earth
Society every time they mention a round-the-world yacht race, and
stretches it to the climate debate.
Which is barmy for two blindingly obvious reasons: first, the
UN’s own climate projections contain a range of outcomes from
harmless to catastrophic, so there is clearly room for debate; and
second, this is an argument about the future not the present, and
you cannot havecertainty about the future.
The BBC bends over backwards to give air time to minority
campaigners on matters such as fracking, genetically modified
crops, and alternative medicine. Biologists who think GM crops are
dangerous, doctors who think homeopathy works and engineers who
think fracking has contaminated aquifers are far rarer than climate
sceptics. Yet Greenpeace and Friends of the Earth spokesmen are
seldom out of Broadcasting House.
So the real reason for the BBC’s double standard becomes clear:
dissent in the direction of more alarm is always encouraged;
dissent in the direction of less alarm is to be suppressed.
I sense that some presenters are growing irritated by their
bosses’ willingness to take orders from the green movement. Others
no doubt justify this bias to themselves by saying that climate
dissenters get plenty of exposure in newspapers. This conveniently
ignores The Guardian’s and The
Independent’s almost comical bias towards alarm on the topic
of climate change. It also ignores the fact that the BBC, funded by
a compulsory poll tax and unregulated by Ofcom, is by far the
dominant source of news for most people in this country, with a
market share that should have had prompted an investigation by the
competition authorities years ago. And it is required by charter to
be balanced.
Incidentally, I have vested interests in energy, too, but I am
not asking for people who disagree with me to be silenced.
Matt Ridley's Blog
- Matt Ridley's profile
- 2180 followers
