Friday, 31 January 2014

China as economic powerhouse

As Chinese NY comes around it is time to comment on an issue that has concerned me for some time: The question of Chinese economic preeminence.

A regular theme in English (US) language media has been that the masses have incorrectly judged that China is now the preeminent economic power. The journalist in such cases often then calmly explains that this is not the case: The US economy is still massively larger than China's.

As a academic within machine learning I am tempted to believe the wisdom of crowds. As a keen observer of international affairs and the related economic statistics, I am sure we should.

The US economy is roughly twice as large than China's when measured on a nominal basis. Measured by PPP the difference is only approximately 80%.  Others have (long ago) mentioned that the calculations involved in the PPP deflator for China are possibly so biased by the fact that they are drawn from the major cities that China may already be the world's largest economy.  But the major point I wish to make is something different: Economic power is not given by an economy's size alone.

What is important in terms of international power is not an economy's size but the ability the regime of that country has to influence others on the basis of its economic might. China's economy may be smaller than the US, but (i) it is much more open to the world, meaning that it is the largest trading partner of many other countries, (ii) it is growing much faster than the US (particularly true in nominal terms as currency appreciation and inflation play their role), and (iii) it is much better fiscal position, permitting it to invest in (and provide loans to) other countries on a major scale.

If economic power is purely by nominal trade volume, China would win. Including some form of horizon weighting for growth only increases its importance. Including the appreciation and inflation make it obscene.

So what is going on? Are reporters simply stupid? Yes and no. There are obvious responses to trying to claim that China is the world's most powerful economy from the above. Most obviously, the US sits in a very privileged position due to its role as the issuer of the principle reserve currency. It also welds disproportional power through its institutional position with the economic system. There is also the red herring of its ability to respond to disasters (yes - it provided much more relief to the Philippines, but so what?).

But more than anything else, the reasons for denying the obvious fact of China's economic preeminience are four-fold:

(i) Economic importance is only one facet of international relations. Many countries are now in the position that their economic trajectory relies more heavily on China than the US (or the EU), but are nonetheless much more beholden to their relations with the US. This can be true because of military or cultural reasons. But the fact that such countries remain principally tied to the US is not because of its economic power.

(ii) Most Americans, Westerners and even many beyond these cultural boundaries do not want to acknowledge China's economic power. Most important in the third group is China itself. Less than two decades ago, a Japanese prime minister noted that Chinese growth was illusionary, located only along its coast. In the past decade, theses that China could only develop until it competed in low wage, low margin industries were legion.

(iii) The emergence of China as a economic powerhouse is no different to the previous claims that Japan or Germany (or France) were inevitably going to surpass the US. Actually this is, in some sense, certainly true. In each of these cases, the economic growth leveled off. Arguments can be made for whether these countries reach US levels of development (per hour it appears so, but per capita it does not). But this misses the point. For Germany or Japan to surpass the US economically required them to triple the per capita GPD of the US. US productivity needed not simply to be equaled but massively surpassed. Both these countries either basically equaled (per hour) or approached (per capita) US productivity. China needs to do neither. It needs simply to develop to a middle income country with productivity well below the US to become to world clearly preeminent economic power. When you listen to rumors of China over investment, remember that it has a capital stock per capita far less than, say, Spain.

(iv) Finally, let us not deny that English language newspapers decry such popular assumptions of Chinese economic development because such articles are conducive to their readership, who want reassurance that the current US dominated system is impervious to change.

It is not.

Tuesday, 21 January 2014

Sweden: Give me gender equality, but don't mention the immigrants

I now live in Sweden, a country which is, to many in the English speaking world, a synonym for decent intelligent social-democratic society.

Of course, there is much truth in this. But there is much falsehood too. As part of bidding farewell to 2013, Sweden's Dagen Nyheter ran its obligatory end of year analyses. This is one of Sweden's high-brow papers. And this included a section on the most important blog (posts) of the year. Unfortunately I read this in paper, so there is no link here. But if I remember correctly a full 7 of the ten were gender related.

Now this may be unsurprising: Sweden is known for its gender equality, and, as these posts were happy to reveal, gender inequality still persists in some shape or form even here. However I must admit to feeling uncomfortable in this concentration of attention.

With the very important exception of females being the victims of violent acts, gender inequality in Sweden is becoming highly nuanced. And five of these blog were clearly unrelated to the violence question (memory accuracy assumed). Yes - women earn less than men. But they chose less remunerative careers. This could reflect societal bias against women, or it could reflect personal choice. In which case enacting social engineering remedies aimed at altering the phenomenon risk 'destroying the village to save it'. If females have certain norms and aims, and gender is socially constructed, re-engineering these norms and aims does not empower the group, it eliminates it.

You may disagree. Or at least disagree that we are at this point. If the latter you are probably correct. But increasingly, gender discrimination in the concrete requires some major prejudicial assumptions of its own. The female interviewer was asked by her male interviewee who wrote her questions: The interviewee would never have asked that of a man! (Why not?) The nobel prize winner was asked if she simplified her writing by a male interviewer: He would never have asked that of a male! (Why not?) Now to some degree this is a matter of prior and posteriori probabilities. But to some degree it also reflects a willingness to prejudicially impose negative motives in order to discover gender prejudice.

What is the problem with this?

The problem is that Sweden has much un-nuanced prejudice. No evidence I have seen indicates that that a women would be likely to be refused a job interview because she was a woman. Nor would her education achievements be dismissed on the basis of her gender. Yet research shows than Swedish citizens with non-European names routinely face such overt discrimination. Stockholm has the most highest educated taxi drivers in the world. Unemployment amongst immigrant communities languishes far above the national average even though they have a higher proportion of university graduates. A political party with roots in extreme Nazi ideology now has members in parliament. My wife teaches at gymnasium (upper high school) and her 'immigrant-named' students talk opening of having to research the ethnicity of an employer before bothering to apply.

Yet in the midst of this, Swedish politicians note that "Sweden is not a racist country." Yet they do concede that sexism still exists.

It is this that makes Swedish preoccupation with nuanced sexism unforgivable. I agree that gender inequality is incomplete. But politics is about focus. Concentrated focus on gender issues is only defensible where they eclipse other issues. That is no longer the case in Sweden.

The reasons for this are both fair and foul. Partly it is because (a) generation(s) of Swedes have been raised to see gender issues as the central political issue of our time.

But it is also because those who identify as 'ethnic Swedes' have little willingness to confront the fact that the social inequality that now exists in Sweden is now emphatically in their favour, male or female. Many view the very existence of non-ethnic Swedes in Sweden as a favour to the latter and refuse to note either the reliance of the Swedish welfare system on the importation of immigrants or the fact that 'ethnic Swedes' are experiencing upward social mobility while immigrants face the reverse (despite their aforementioned education advantages, many of which are unrecognised).

But perhaps the most plausible explanation of the media's preoccupation with gender issues is the most mundane: It permits them to preoccupy themselves with a problem that can then be used as a means to self-congratulation. For while sexism occurs in Sweden, we can easily show that it does so less than in those god-forsaken foreign lands. And thus the Swedish media's readers leave satisfied both that they have touched a problem and that their society is deservedly lionised in its attempts to remedy it. The ills that actually cause such damage here, though, remain unaddressed.

Friday, 17 January 2014

Economist: Bad time to be stupid

The economist informs us that technology is placing increasing pressure on lower income jobs:
One recent study by academics at Oxford University suggests that 47% of today’s jobs could be automated in the next two decades. ... If this analysis is halfway correct, the social effects will be huge. Many of the jobs most at risk are lower down the ladder (logistics, haulage), whereas the skills that are least vulnerable to automation (creativity, managerial expertise) tend to be higher up, so median wages are likely to remain stagnant for some time and income gaps are likely to widen.
No doubt much of this is true. Many current jobs will be automated, and many of these jobs will be 'lower down the ladder'. But is it true that these jobs that are most at risk are those that are least vulnerable to automation, and it is because of this that such 'lower' jobs face automation?

I am not convinced. The reason for this is that many of the jobs most able to be automated by the latest machine learning and automation technologies are high prestige, high wage positions. Automated medical diagnostics is now able to outperform your local GP, as well as many of their specialized colleagues.  Text based search can outperform any lawyer in searching for precedents in massive libraries of case histories. Managerial expertise, particularly the out-sourced consultant variety, is incredibly vulnerable to automated data analysis. The political class is arguably obsolete (at least in democracies) in an era where instant communication overcomes the obstacle of distance that plagued former forms of direct democracy (and no - it's not insane). Let us not even begin to discuss the ability of automated algorithms to render suboptimal your human financial experts...

The above paragraph may cause some cognitive dissonance. It is unlikely that technology will cause major job losses among doctors, lawyers, management consultants, politicians and financial experts. (Well, OK, maybe lawyers - supply and demand are finally making things are start to look grim for this formerly cosseted profession.) But it important to note that the reason why this is unlikely has nothing to do with technological progress. Rather, these professionals have excellent institutional and cultural protection from the technological trends that are buffeting lower income individuals.

GPs are not going to be replaced as our front line medical because the medical establishment would raise hell. Politicians and political parties are very unlikely to approve the investigation and validation of alternative governance methods where these would render them superfluous. No mention has been made of permitting defendants to have access to legal search algorithms when defending themselves. Financial mandarins in the major investment banks and financial centers continue to have sufficient market power to award themselves multimillion bonuses.

So the issue is not that neutral technology is about to cause major disruptions that will happen to fall disproportionately on lower income professions. Rather it is that the lower income professions will disproportionately feel the effects of these disruptions because powerful forces are ensuring that they bear this brunt.

In this context, it is unfortunate that the economist claims that: 
Anger about rising inequality is bound to grow, but politicians will find it hard to address the problem. Shunning progress would be as futile now as the Luddites’ protests against mechanised looms were in the 1810s, because any country that tried to stop would be left behind by competitors eager to embrace new technology. The freedom to raise taxes on the rich to punitive levels will be similarly constrained by the mobility of capital and highly skilled labour.
No. Politicians would find it relatively easy to let all sectors of the economy face the same tumultuous forces that the economist warns about. But there is very little interest in doing so.

What does the economist suggest instead? Education, of course.
schools themselves need to be changed, to foster the creativity that humans will need to set them apart from computers. There should be less rote-learning and more critical thinking. Technology itself will help, whether through MOOCs (massive open online courses) or even video games that simulate the skills needed for work.
Ignore the fact that most of the professions that supposedly avoid the technology tsunami are doing so through exploiting privileged positions in the economic system. Ignore also that nobody is really sure what critical thinking skills are or why they are important. Also ignore that MOOCs have abysmal graduation rates. 

Fortunately, the economist has an answer to any failure of their remedy to address the problem:
Yet however well people are taught, their abilities will remain unequal, and in a world which is increasingly polarised economically, many will find their job prospects dimmed and wages squeezed.
Ah. It turns out that those who will lose out in the coming dislocation are probably simply stupid...