Arguing about infographics with Galileo

I recently signed up for data vizualization guru Edward Tufte’s one-day course, mostly as an attempt to burnish my chart-geek credentials. I got rather more than I bargained for: the course was not really about how to make better infographics, or even about how to give business presentations (though both topics were addressed). It was more of a long ramble through Tufte’s mind and his obsessions–which are not so much data as information more broadly, and not simply vizualization but communication more generally.

This finally became clear to me during an enjoyable but initially somewhat puzzling discussion of his recent visits to his doctor. It seemed like a stream-of-consciousness digression at first, but then it became clear he was thinking through a serious issue: how to best communicate critical information in a stressful setting. (Tufte’s tip: write out all your medical concerns at home before going to the doctor, then hand over the document at the appointment). If you can’t get your doctor to understand what you need, he was implicitly saying, then what hope do you have of getting people to understand anything less important?

But while Tufte’s concerns are not limited to charts, he has spent a lifetime thinking through what he called the “perennial” problem of how to represent a multidimensional world in the two dimensions of the page or screen. At the end of the day, he pulled out a first edition of Galileo Galilei to show how the great minds of the past had grappled with the same issues. He rhapsodized over Galileo’s tiny, in-line sketches of Saturn, which clearly inspired his own advocacy of “sparklines” (tiny charts embedded in text at the same size as the text), as well as some beautifully precise illustrations of sunspots.

Tufte also showed an engraved portrait of Galileo, in which he has appears with an engaging smile, and called him “funny, bright-eyed…a bit like Richard Feynman.” It was clear that he felt he knew Galileo as a person through his work, and felt a deep connection to someone who had been working on the same problems: the rigorous collection of data and its careful presentation.

I was particularly sensitive to this dynamic because I had just finished reading Oliver Sacks’ loving discussion of the great early chemist Humphrey Davy, contained in his posthumous essay collection Everything in its Place: First Loves and Last Tales:

Humphry Davy was a boyhood hero to virtually everyone interested in chemistry or science in my generation. We all knew and repeated his famous experiments, imagining ourselves in his place. Davy himself had had such ideal companions in his youth, particularly Newton and Lavoisier. Newton, for him, was a sort of god; but Lavoisier was closer, more like a father with whom he could talk, agree, disagree. His own first essay, which Beddoes had published, while taking strong issue with Lavoisier, was in effect a dialogue with him. …

When I came to write my first book, Migraine, in 1967, I was stimulated by the nature of the malady and by encounters with my patients, but equally, and crucially, by an “old” book on the subject, Edward Liveing’s Megrim, written in the 1870s. I took this book out of the rarely entered historical section of the medical school library and read it, cover to cover, in a sort of rapture. I reread it many times for six months, and I got to know Liveing extremely well. His presence and his way of thinking were continually with me. My prolonged encounter with Liveing was crucial for the generation of my own thoughts and book. It was just such an encounter with Humphry Davy, when I was twelve, that had originally confirmed me on the path to science.

I do not think my experience is unique. Many scientists, no less than poets or artists, have a living relation to the past, not just an abstract sense of history and tradition but a feeling of companions and predecessors, ancestors with whom they enjoy a sort of implicit dialogue.

Tufte’s on-stage dialogue with a four-hundred-year-old work by Galileo was, among other things, a demonstration of how effective it is to teach and learn about science in a historical way, as a sequence of personalities, problems, and arguments.

The difference between the new and old Cold Wars

The deteriorating US relationship with China is more and more frequently being called a “cold war” and compared to the long-lasting rivalry with the Soviet Union. Even before the latest breakdown in US-China trade talks, it seemed like negotiations would at best produce a settlement of specific economic issues, and leave the broader relationship pretty frosty.

But a true “cold war” with China, if the administration does decide to go down that route, is in practice unlikely to play out in the same way as the US-USSR confrontation. This is for the simple reason that the US and Chinese economies are already intertwined to an enormous degree. By contrast, the US basically did not trade at all with the Soviet Union, and trade with its successor states has remained quite minimal. I find the simple chart below to quite striking. So for the US, a cold war with China is much more economically risky than isolating the Soviet Union ever was.

Another way to look at the US trading relationship with China is not to compare it not with historic rivals, but with other major trade blocs. It is a simple but important point that the majority of US trade is with friends and allies: Canada and Mexico, Europe, and the Asian democracies of Japan, South Korea, and Taiwan. The US has allies in the Middle East, but these alliances can be somewhat uncomfortable ones given the autocratic nature of the governments (Turkey, Egypt, Saudi Arabia). In any case the US economic integration with the Middle East is really quite small. China stands out for how it is, at the same time, both a political rival to and economically integrated with the US.

So from one perspective, the US economic relationship with China is too large and important to be casually endangered, even if it does need repairing. From another perspective, China is too different politically from the US to be permitted this degree of economic integration. I don’t know which perspective will end up dominating.

What I’ve been listening to lately

  • Frank Kimbrough – Monk’s Dreams: The Complete Compositions of Thelonious Sphere Monk. Obviously, a lot of Monk: all 70 compositions are given full and respectful readings by a jazz quartet, for more than five hours of music. Less obviously, it’s also a lot of bass saxophone. Scott Robinson, who occupies the horn chair in the quartet, is almost as promiscuous in his multi-instrumentalism as Anthony Braxton, and like Braxton shares a love for the low-end instruments. Hearing the supposedly cumbersome bass sax lightly dance through the quirky angles of Monk’s tunes is one of the highlights of this tribute. 
  • John Luther Adams – Four Thousand Holes. A gorgeous piece of minimalist composition, which is indeed built from very simple elements: piano, percussion, a bit of electronics. Adams is one of my favorite contemporary composers. 
  • Melba Liston – Melba Liston And Her ‘Bones. The first (perhaps only?) prominent female trombonist in jazz, Liston was also a gifted arranger and in her later career worked extensively with the great Randy Weston. This package collects some of her work from the mid-50s, and her interesting arrangements lift the material above other mainstream jazz of the period. 
  • Keith Hudson – Playing It Cool. An amazing piece of dark, rhythmic dub experimentalism from 1981. Almost everything I’ve heard by Hudson is absolutely essential, funky, compelling and strange in equal measures. 
  • Charlie Haden – Not In Our Name. Haden’s 1968 Liberation Music Orchestra is one of the classics of radical 60s jazz, but (whisper it) I actually like this 2004 follow-up recording even better (the orchestra itself is entirely different, sharing only Haden and pianist/arranger Carla with the original incarnation). Jazz compositions are intermixed with Bley’s transformations of quintessential American musical pieces such as Amazing Grace and bits of Barber’s Adagio for Strings and Dvorak’s New World Symphony. 

Yasheng Huang on historic human capital in China and India

The always interesting MIT professor Yasheng Huang has done a long podcast for the University of Pennsylvania’s China series. He starts off by criticizing people who compare China and India to argue that China’s authoritarian state capitalism is better for growth than democracy, and dives into economic history to explain why that isn’t right (quotes are from my notes, lightly edited for readability):

If you want to make an apples-to-apples China and India comparison, you need to control for other differences between the two countries. And the basic thing you need to control for is the quality and quantity of human capital. I would argue that unambiguously, China has done a thousand times better than India in terms of human capital development: public health, public education. Historically speaking, in part because of the exam system, China has always had a very strong tradition of literacy, being able to read and write. There is some evidence to suggest that China’s mass literacy in the 17th and 18thh centuries was comparable to that in Britain. This is going way back. I do see that as a huge strength.

A lot of the growth differences between India and China and India are really explained by that. So there is a fundamental attribution error that many people have committed. When they look at the differences between China and India they say, one is a democracy and one is an authoritarian system, one has better GDP growth and the other has worse. Little do they know that there are other differences. It’s these other differences that explain the growth difference between China and India. I would say that human capital explains 80% of the differences. Maybe we should take that more seriously.

China has always had something behind its back to have good, solid economic performance. Even in the 16th and 17th centuries they had pretty good performance by the standard of that time. In that sense, I’m not a free market fundamentalist. I see the state as being absolutely critical in building the human capital base. This is what the Chinese did historically, and also what the Chinese did during the Communist period, and also what the Chinese state is doing today. For that I give them an A-plus, I celebrate their achievements.

Huang said he is working on a new book that will investigate these historical foundations to China’s growth today (and also said he is working on an updated edition of his well-known bookCapitalism with Chinese Characteristics to incorporate post-2008 events).

There is perhaps a tinge of motivated reasoning here, as Huang is clearly looking for ways to explain China’s economic growth miracle without giving the credit to Chinese state capitalism. But I’m sympathetic to the idea that pre-1949 China, rather than being the backward feudal hellhole of Communist propaganda, was in fact pretty well equipped for modern economic growth–at least, once it could manage to put an end to foreign invasion, civil war, and aggressively backward government policy. Indeed, Huang’s arguments echo points made by the great economic historian Dwight Perkins, who also emphasized the importance of pre-20th-century China’s functional bureaucracy and solid education:

China’s capital city had a population of over one million people as early as the Song Dynasty, if not before, and supplying such a city required tens of thousands of merchants, transport workers and the like. Commerce on this scale requires records, and to use records an individual must be able to read at least numbers and some characters.

We do not yet have a reliable estimate of the level of literacy in 19th century China, but among males a basic level of literacy could have been as high as 30-45%. Among the highest income 5-10% of the population, literacy must have been nearly universal, and for many at this level literacy went way beyond basic.

A relatively high level of literacy by pre-modern standards did not lead to sustained economic growth prior to the 20th century, but it did lay the foundation for the creation of a modern high-quality education system, at least when one compares the education system that existed in China in 1949 with what one found in much of the developing world on the eve of that world’s attainment of independence from colonial domination.

(The source is Perkins’ The Economic Transformation of China, pp. 7-8.)

The challenge, of course, is to produce more rigorous measures of historic human capital and educational achievement that could test these impressions. It will be interesting to see what Huang comes up with.

The Belt and Road is about domestic interest groups, not development

Andreas Fulda on Twitter pointed out a useful new piece on China’s controversial Belt and Road Initiative by CSIS research Mark Akpaninyie. It seems that Mark and I have been thinking along similar lines, and the resulting online exchange helped me clarify my thinking.

It’s become increasingly clear that the “debt-trap diplomacy” meme started by Indian commentator Brahma Chellaney is not an accurate description of how the Belt and Road actually operates, despite the fervent embrace of this idea by China hawks. Basically, China is not actually organized enough to come up with such a clever and nefarious plan, and there is no evidence that there is a deliberate strategy to trap other countries in debt. A detailed examination of debt transactions by Rhodium Group also found that in many cases borrowers were able to get China to write off or renegotiate their loans.

The flaw in the debt-trap diplomacy theory, and with many other analyses, is that it mistakes the Belt and Road for a for a “highly centralized and coordinated” initiative. In reality, it is more of a slogan attached to the decentralized actions of state-owned enterprises and banks. Here is how Mark Akpaninyie describes it:

Little evidence actually suggests that Beijing coordinates a unified strategy to lure the developing world into unsustainable debt.

Instead of a state-led strategy, Chinese firms — motivated by profit and abetted by a toxic combination of bureaucratic disorganization, incompetence, and negligence at the state level — have exploited poor nations, which are dependent on cheap, and sometimes bad, loans. These companies, knowingly or unknowingly, persuade countries to pursue projects where benefits to the firms far outpace the benefits of the host nation. …

This practice does not trap recipient countries into taking on unsustainable debt. Instead, it allows Chinese companies to profit from often crooked deals building much-needed infrastructure in some of the world’s poorest countries, exploiting the undersupply of financing and these countries’ appetite for infrastructure projects.

The broader point here is that looking at the Belt and Road through the lens of “grand strategy” or “geopolitics,” as so many commentators do, or even portraying it as some kind of new philosophy of economic development, is quite misleading. All of these grand concepts are justifications invented after the fact for a pattern of actions that was already well underway before Xi Jinping made his 2013 speech about the Belt and Road.

The Belt and Road is really the expansion of a specific part of China’s domestic political economy to the rest of the world. That is the nexus between state-owned contractors and state-owned banks, which formed in the domestic infrastructure building spree construction that began after the 2008 global financial crisis (and has not yet ended).

Local governments discovered they could borrow basically without limit to fund infrastructure projects, and despite many predictions of doom, those debts have not yet collapsed. The lesson China has learned is that debt is free and that Western criticisms of excessive infrastructure investment are nonsense, so there is never any downside to borrowing to build more infrastructure. China’s infrastructure-building complex, facing diminishing returns domestically, is now applying that lesson to the whole world.

In Belt and Road projects, foreign countries simply take the place of Chinese local governments in this model (those who detect a neo-imperial vibe around the Belt and Road are, in this sense, onto something). Even the players are the same. In the 1990s, China Development Bank helped invent the local-government financing vehicle structure that underpinned the massive domestic infrastructure boom. Now, China Development Bank is one of the biggest lenders for overseas construction projects.

Those who defend the Belt and Road against the charge of debt-trap diplomacy are technically correct. But those same defenders also tend to portray the lack of competitive tenders and over-reliance on Chinese construction companies in Belt and Road projects as “problems” that detract from the initiative’s promise. They miss the central role of the SOE infrastructure-complex interest group in driving the Belt and Road. Structures that funnel projects funded by state banks to Chinese SOEs aren’t “problems” from China’s perspective–they are the whole point.

The fact that this model was dubbed the “Belt and Road Initiative” and turned into a national grand strategy by Xi Jinping effectively gave the SOE infrastructure complex carte blanche to pursue whatever projects they can get away with. These projects were no longer just money-makers for SOEs, but became a way to advance China’s national grand strategy–thereby immunizing them from criticism and scrutiny.

None of this means that the Belt and Road will not change or evolve. But I suspect that the trajectory it will follow will be similar to that followed by local-government infrastructure projects in recent years. The central government does actually worry about excess debt and bad projects, and so the building and funding of infrastructure have become gradually subject to more discipline and central scrutiny. But this has been done in a way that does not shock the entrenched domestic interest groups, and overall economic growth, too badly.

After last week’s forum, it does look like the Belt and Road is also on the way to becoming a bit more organized. But given the driving role that domestic interest groups have always played, hopes that it will be turn into a benevolent and technocratic global economic-development program are going to be disappointed.

William James on the value of doctorates and diplomas

Greg Ip at the WSJ has a nice piece responding to the ruckus over the nominations of Stephen Moore and Herman Cain to serve on the Federal Reserve Board. It’s obvious from the Fed’s own history that the mockery of Moore for not having published peer-reviewed journal articles, or not having a Ph.D. in economics, quite misses the point. As Greg nicely puts it, the real question to ask about someone who is may need to make economic policy decisions is whether they are a disciplined thinker, not whether they have a certain credential.

By coincidence, I also recently read an essay by William James entitled “The Ph.D. Octopus,” originally published in the Harvard Monthly in March 1903 (it was reprinted in his essay collection Memories and Studies which is out of copyright and freely available). Some of James’ sentiments still ring quite true:

America is thus as a nation rapidly drifting towards a state of things in which no man of science or letters will be accounted respectable unless some kind of badge or diploma is stamped upon him, and in which bare personality will be a mark of outcast estate. It seems to me high time to rouse ourselves to consciousness, and to cast a critical eye upon this decidedly grotesque tendency.

James worried that the institutionalization of graduate degrees, and in particular their use by employers to screen potential hires, would cause all kinds of negative consequences:

To interfere with the free development of talent, to obstruct the natural play of supply and demand in the teaching profession, to foster academic snobbery by the prestige of certain privileged institutions, to transfer accredited value from essential manhood to an outward badge, to blight hopes and promote invidious sentiments, to divert the attention of aspiring youth from direct dealings with truth to the passing of examinations…

James was deeply aware of the tension between universities’ avowed mission of free intellectual inquiry and their economic function as producers of credentials, and hoped that the former would discipline the latter:

Our universities at least should never cease to regard themselves as the jealous custodians of personal and spiritual spontaneity. They are indeed its only organized and recognized custodians in America today. They ought to guard against contributing to the increase of officialism and snobbery and insincerity as against a pestilence; they ought to keep truth and disinterested labor always in the foreground, treat degrees as secondary incidents, and in season and out of season make it plain that what they live for is to help men’s souls, and not to decorate their persons with diplomas.


Rediscovering the importance of export discipline

The new IMF working paper on industrial policy, by Reda Cherif and Fuad Hasanov, has gotten a lot of notice, and indeed it is very clear, comprehensive, and useful. But for anyone who has already done some reading on the history of successful Asian economies, particularly Taiwan and South Korea, it is not exactly surprising stuff. Here for instance is their quick summary of the key characteristics of these economies’ successful industrial policies:

  • Intervene to create new capabilities in sophisticated industries: Pursue policies to steer the factors of production into technologically sophisticated tradable industries beyond the current capabilities to swiftly catch up with the technological frontier.
  • Export, export, export: A focus on export orientation as any new industrial product was expected to be exported right away with the use of market signals from the export market as a feedback for accountability. As conditions changed, both the state and the firms adapted fast.
  • Cutthroat competition (at home and abroad) and strict accountability: No support was given unconditionally although performance assessment was not necessarily based on short term profits. While specific industries may get support, intense competition among domestic firms was highly encouraged in domestic and international markets.

The combination of a focus on exports with tough competition sounds a lot like what Joe Studwell, in his 2013 book How Asia Works (which is not cited in the IMF paper’s bibliography), called “export discipline.” His explanation is clearer and punchier:

Governments in all the major economies of east Asia tried at some stage to nurture domestic manufacturers. That those in north-east Asia succeeded, while those in south-east Asia failed miserably, turned on a small number of policy differences. By far the most important of these was the presence – or absence – of what I call ‘export discipline’.

This term refers to a policy of continually testing and benchmarking domestic manufacturers that are given subsidies and market protection by forcing them to export their goods and hence face global competition. It is their level of exports that reveals whether they merit state support or not. …

Where export discipline has not been present, development policy has become a game of charades, with local firms able to pretend that they have been achieving world-class standards without having to prove it in the global market place. In south-east Asia, the energies of entrepreneurs were directed towards fooling politicians rather than exporting.

I would still recommend Chapter 2 of How Asia Works as the definitive comparison of successful and unsuccessful industrial policies in Asia.

The point of such a comparison is to move beyond sterile debates over whether industrial policy can ever work, since in fact basically all countries have some kind of policy for promoting particular industries. As Cherif and Hasanov put it, “The key question is, if many countries have been conducting industrial policy anyway, what should the right way to do this be.” The presence or absence of export discipline should be a useful way to evaluate whether industrial policy is likely to be successful.

Even within Asia this lesson is not as widely appreciated as it perhaps could be. For instance, former Chinese finance minister Lou Jiwei recently made a surprisingly harsh public criticism of Made In China 2025 (for which he has apparently been forced into early retirement). He called it a waste of taxpayers’ money and an unwarranted intrusion of government: “those industries are not predictable and the government should not have thought it had the ability to predict what is not foreseeable.”

While I have a lot of respect for Lou, I’m not sure this is the strongest criticism of Made in China 2025. It’s not clear that “the market” would necessarily pick different industries as being desirable to invest in now: the ideas that people have about what technologies are going to be important in the future don’t seem to be that different across the public and private sectors. The Chinese government have have a plan to promote artificial intelligence, but private venture capital firms are also throwing plenty of money at that sector as well. Semiconductors are one of the key sectors targeted by The Made In China 2025, and I don’t think many people are seriously arguing that semiconductors won’t be important in the future.

This is not to say that venture capital investors are necessarily going to be right about the future either, just that both government officials and venture investors can read the same things and are influenced by the same conventional wisdom. This point is not original to me: I picked it up from Brad DeLong’s 2010 book with Stephen Cohen, The End of Influence:

Americans like to say scornfully that industrial policy is about “governments picking winners.” Picking winner industries is not that hard—even for governments. Most countries trying to climb the ladder of quality and industrial sophistication through selective promotion compiled pretty much the same lists at the same time. Even at the leading edge of the technological frontier, the industries that governments are tempted to promote are largely the same ones picked by the analysts and brokers at investment firms such as Merrill Lynch, Nomura, or Rothschild’s.  …

Picking “winner industries” is not the hard part; winning is. It is difficult to create actual winners, companies that develop into successful competitors.

And that, of course, is where export discipline comes in.