Daniel Lemire's blog

Everyone knows that publicly funded education is good. Right? Wait! Why?

• “Schools have substantial non-financial benefits.” This argument assumes that people who forgo schooling are uneducated. It is weaker in the Wikipedia era. Kids are naturally curious, and they now have access to unlimited and inexpensive information. And this same argument could be used to justify free Internet for all, which would be considerably cheaper than free schools.
• “If it costed hundreds of thousands of dollars to complete a Ph.D., nobody would do it.” Assuming that there is any demand at all for Ph.D.s, people with a Ph.D. would receive much higher salaries if fewer people have them. These higher salaries would entice more students to complete a Ph.D.
• “Poor people are unable to get an education without government funding.” Students who can borrow the money, ought to be willing to do so if the expected return on their education investment far exceeds the interest rates charged by the bank or a private investor. True: Some students who show little promise, and have few ressources, would be unable to get an college degrees. Is that fair? Is it fair that only the most promising engineers get a job with Google? Is it fair that my wife is more beautiful than yours? Is it fair that kids go hungry in the richest country in the world?
• “To make our corporations more competitive.” By funding schools, governments entice more students to study which artificially boosts the supply of graduates. In turn, this lowers the salaries of these same graduates. Corporations benefit from these lower wages while they only contribute a small fraction of the cost. Therefore, public schools are equivalent to subsidizing corporations. Any country that would stop funding higher education without a corresponding immigration policy, would see a rise in the wages of college graduates. This would be unfavorable to corporations which rely on college degrees to select employees.  But corporations do not have to hire college graduates. Most corporate jobs are a form of paper pushing. They could easily replace college degrees with less expensive certifications.

So,  why should the public fund schools?

Further reading: How and Why Government, Universities, and Industry Create Domestic Labor Shortages of Scientists and High-Tech Workers, A World Without Public School, What If Public Schools Were Abolished? and Self-interest and public funding of education.

Disclosure: I work for a public university. My kids attend a public school.

A programming language is Turing complete if it equivalent to a Turing machine. In practice, it means that any algorithm can be implemented. Most programming languages are Turing complete, including SQL, ECMAScript, Java, C, and so on.

HTML5 + CSS3 is also Turing complete because it can be used to program a Rule 110 automaton.

Source: Thanks to Jakob Voß for the pointer.

The Luddite fallacy says that innovation destroys jobs. It is believed to be a fallacy because increased productivity causes prices to fall, which then boosts demand which then creates new jobs. While locally or temporarily, there might be economic growth without job creation, economists believe that job growth will always follow. In effect, it represents the belief that technological innovation is always good economically.

This is a central dogma among economists. Why? Because jobless growth would destroy mass market economies. Indeed, jobs are the primary mean by which money flows back from the corporations to the consumers. Without jobs, consumers leave the market, effectively destroying it. Should the Luddite fallacy fail—if labour-saving technologies did increase unemployment—we would be pushed into some form of communism where the governments would need to either artificially create jobs, or provide direct financial support to the population.

And it is not necessary for the Luddite fallacy to fail entirely to have problems. Imagine that as we innovate, maybe through better machine learning and robotics, we keep losing 2% of all jobs every year. For every 100 jobs destroyed, only 98 new jobs are created. Within a few short years, we would be left with a sizeable fraction of the population which has left the job market entirely, and another fraction which is either unemployed or underemployed.

As bank tellers, cashiers, secretaries, and middle managers are being replaced by software and robotics, the conventional wisdom is that new jobs are created. Maybe people become robotics technicians or solar panel experts. However, how many futuristic jobs are really created? Many of the jobs created in new companies are the same jobs that disappeared: administrative assistant, accountant, manager, engineer, and so on. Certainly, Facebook is a futuristic service. But while it serves hundreds of millions of users, it only has 2000 employees.

In his latest TED talk, Bill Gates hinted that the future looks bleak for governments. Health costs are rising faster than government revenues, even if you assume that the economy will do well. Thus, it is entirely possible that we will need to drastically reduce the cost of healthcare and education. Automation might be the only path forward which would not sacrifice our quality of life. Will we be closing schools and sending the kids to Khan Academy? We may have to automate much of the medical testing and diagnosis. Robotics might be needed to help support an aging population. All these great innovations might be necessary, and they may also lead to further job destruction (or the equivalent lack of job creation).

To make things worse, much of the recent gains from technology are nonmonetary. They fall outside the realm of economics. For example, while journalists have been losing their jobs, I have had access to better content than ever, through blogs and e-books, mostly for free or for little money. Recently, the giant free porn site RedTube has been killing much of the online porn industry: it offers unlimited porn for free.

Should the Luddite fallacy fail, there is an alternative to communism. We should be entering into the age of the 4-hour workweek. We have been investing our productivity gains into more production, which has lead to more jobs. But what if the cycle breaks at some point in the future? The solution might be to redefine “work”. Many bureaucrats only create work for themselves and others. Most of accounting, for example, could be automated. Researchers famously publish papers only for the sake publishing papers. Instead, we should call on accountants or researchers a few hours, here and there, to handle the interesting and difficult problems. Employers, instead of buying 40 hours of someone’s time, would buy the option of calling on the individual in times of need. We would all work far fewer hours, we would preserve the market economies and we would be happier. To make this possible, we need a deep cultural change. How do you feel about people who work only a few hours a week, or a month? My impression is that we look down on anyone who works less than 30 hours a week. Presumably, we inherited this prejudice from the industrial age. We need to let go.

Further reading: The Lights in the Tunnel by Martin Ford.

When designing an information system, a piece of software or a law, experts work from a model. This model must have boundaries. When these boundaries are violated, life may become difficult:

• After over two years in lock-out, the journalists of the largest Montreal newspaper decided to bow to their bosses and accept whatever they could. What happened? In Quebec, we have a strong anti-scab law. In theory, therefore, you cannot publish a newspaper without your newsroom. Except, of course, that the law assumes work happens in the offices. In practice, anyone can work for a newspaper from home, sending documents by email. The boundary of the work model in this case is the (physical) office. Entrepreneurs are smart enough to realize that they do not need the conventional journalists, working in the newsroom. Alas, consumers are now realizing that they do not need the newspapers in the first place. We have broken models all the way up.
• Email was designed to work within a small and  non-commercial network. It was never meant for the world we live in. And I attribute much of the rise of social networks (e.g., Facebook) to the need to recreate these small non-commercial networks. The irony, of course, is that people who are investing in Facebook right now tend to forget that the boundaries of the Facebook model have much to do with avoiding spam.
• Entity-relationship models were designed in an era when corporations had few databases which were relatively small and simple. Moreover, you could isolate various databases: there was no need for your databases to ever be compatible with the databases of another company.   Today, many corporations have databases with hundreds of attributes, all designed by different experts. The semantics can vary widely within the corporation. What is worse is that you must often reconcile the semantics across several companies inside supply chains. The net result is that if  you wanted an actual entity-relationship diagram which represents what is actually going on, it would never fit in anyone’s head. We have done much work on schema evolution and data integration but we fall short of solving the problems without pain.

The boundaries of our models are constantly being broken. These broken pipes represent vast opportunities. A new CEO who wants to turn a company around should look for breached boundaries in the company’s model. Competitors should look for broken water pipes as a way to identify soft targets.

I believe this calls for an entirely new field of expertise. Some people should become experts are recognizing broken models.

Here is my favorite broken model: E-commerce is pathetic and inefficient because it is based on the pretense that we need stores. Like we do not need newspapers (just news), we do not need stores (just products). In the nineties, people were talking about software agents that would go out and shop for us. Why do I have to do so much manual labor to find the best product and the best price? Sometimes, it almost looks like sabotage. Consider this scenario. You go see your doctor. Instead of handing you out a prescription that you will need to present in some drug store, the doctor just tells you “the medicine will be delivered to your house tomorrow morning, it will be automatically debited from your insurer’s account”. See? No store. Ironically, the future of e-commerce might be to do away with stores entirely. Oh! And it won’t be invented by people running e-commerce stores.

Back in 2005, Shirky argued that the Social Web offered an alternative to organizations. Working collaboratively has never been easier. And the innovation is ongoing: collaborative real-time editors (e.g., piratepad) are amazing. GitHub has taken collaborative software production to a whole new level by allowing anyone anywhere to fork any project in seconds. Doodle makes scheduling meetings painless. Silly things like raising funding can be done collaboratively: LunaTik raised a million dollars to design an iPod-watch using Kickstarter.

Each and everyone of these innovations directly compete against one of the services offered by typical organizations. It is not merely that the technology is replacing people (e.g., the office suite replacing secretaries). Instead, the very core function of organizations is challenged: organizing people toward common goals, animating the office life. But not all organizations are equally challenged. Facebook appears to be doing well. Meanwhile, the Egyptian government had a rough time recently.

How can you tell which organization is threatened? I believe that Paquet puts it well:

The internet is looking more and more like one huge machine for disintermediating the clueless.

Paquet was referring to the Gervais principle. In the Gervais model, organizations are made of three types of individuals:

• Workers who exchange their labor against a monetary compensation. They include the creative types, the engineers, the writers, and so on.
• Facilitators (sometimes called the clueless) who produce nothing tangible, but keep the organization running by staffing the hierarchy. Middle managers belong to this class.
• Leaders who run the corporations. They are usually very well compensated. They are interfaces between reality and the life of the organization. They often take the difficult decisions.

One of the useful function of corporations is to create jobs, many jobs. In mature organizations, the clueless can easily form the majority of the employees. However, as the cost of organizing people falls, the need for employees to support the organization also falls. How many people in the average corporation are there to support random indirect functions which have been already automated by the Social Web? Consider Facebook. While Facebook is thought to be worth over $40 billion, it has only around 2000 employees. Normally, a company with nearly unlimited funding would be massively hiring right now: there should be countless middle managers, administrative assistants and the like.

Social Media won’t help overthrow tyrans or CEOs. But it may get rid of the evil middle managers and their assistants. When I read about what motivates the uprisings in the Arab countries like Egypt, it seems that the younger generation is upset with the bureaucracy which makes entrepreneurship difficult. You end up with a weak economy, where all the good jobs are in the ranks of the government or of some large corporations. Yet these young arabs are extremely well informed. They are clever and smart. And they know how to organize themselves very efficiently. They are tired of paying rent to the bureaucrats.

So, it could be that a new tyran will rule Egypt next year. But what I will be paying attention to, is the new Egyptian bureaucracy. The survival of the new government might depend on letting the younger generation organize itself, using social media, to create economic value. True democracy might not come. But I hope for a more agile and open government.

Further reading: Rushkoff has many relevant books, I recommend Program or be Programmed. I loved The Starfish and the Spider: The Unstoppable Power of Leaderless Organizations by Brafman and Beckstrom.

From my experience,  the quality and the quantity of the scientific research articles from China has been increasing dramatically in the last five years. To verify this impression, I crunched some numbers. In the general field where I am most active (Information Systems), China dominates in the total number of research articles published whereas it did not even register before 2003:

Moreover, while Chinese articles are still less cited than American articles, China is catching up. Back in 2005, the average American research article was cited three times as much as the Chinese article. The ratio was 1.75 in 2009.

Conclusion: Whatever policies China is putting in place, it has a dramatic effect on its scientific output.

The case that education pays is often made by comparing the income of people who graduated against the people who did not. The result is compelling.

But there are numerous problems with such naive statistical analysis:

• Kids from wealthy families are much more likely to attend college. Moreover, we expect kids with wealthy parents to earn more than their peers irrespective of their education. So you have to compare the income of individuals with similar backgrounds.
• While I have apparently the same economic background of many of my peers, I pursued a Ph.D. because I was ambitious. Meanwhile, ambition is probably a great predictor for income, irrespective of your level of education. This personal drive is difficult to measure statistically, but you should at least compare individuals with similar expression of ambition (such as grades in high school).
• Even with the same ambition and background, kids have different genetic material. Thus, not only do we want to compare kids with the same background, and the same internal drive, but we must also compare individuals with similar genes.
• Is it fair to compare the yearly salary of a lawyer who has been working 60 hours a week, sacrificing his family and personal life, with that of a secretary working 40 hours a week? Minimally, you should compare the hourly rate.
• Going to college is costly, both in a loss of potential income and in tuition. The cost of the education must be deducted from future income. This explains why earning a Ph.D. is usually a bad financial investment.
• Going to college is costly, both in a loss of potential income and in tuition. The cost of the education must be deducted from future income. This explains why earning a Ph.D. is usually a bad financial investment.
• Arum and Roksa careful document in Academically Adrift that students are not really learning much in college.

Nevertheless, the historical value of education as an investment has been great, even though most naive studies overestimate its value.

But one critical error remains: Can you predict the future based on the past, by naively assuming that the past will be like the future? My bank still claims in its brochures that I can expect yearly returns of 8% on my investments. They have nice plots showing how much investing a little bit right now is enough to become a millionaire in 30 years. And indeed, the annual return on investment from 1950 to today is around 7%. So anyone who invested massively in the 1950s is probably wealthy right now. But the return on investment in the first decade on the twenty-first century has been negative. Some people have made a fortune with their investments lately, but most people have lost money! Is it safe to predict that, eventually, the 1950s will come back? Nobody knows.

Maybe the job market is different? Maybe the market for young college graduates is relatively static and easy to predict? No. The number of university graduates has doubled in the West since 1990, and tripled in Asia. We are undergoing a massive transformation. Brown et al. (2008) warn us that the fast expanding supply of highly trained graduates, both locally and worldwide, will soon push the wages down:

Investment in high skills will not create the high skilled, high waged economy of the future.

(…) the competition for good, middle-class jobs is now a worldwide competition, an auction for cut-priced brainpower, fueled by an explosion of higher education across the world and a fundamental power shift in favor of corporate bosses and emerging economies such as China and India. These drivers of the new global high-skill, low-wage workforce threaten the livelihoods of millions of American workers and their families. Fighting for a dwindling supply of good jobs will compel Americans to devote more time, money and effort to set themselves apart in a bare-knuckle competition that will leave many disappointed.

With the financial market, the solution is to change our expectations: you risk becoming very unhappy if you expect a rate of return of 8%. Similarly, I think we will soon need to change our expectations regarding degrees. More and more smart graduates will be unemployed, or underemployed. They will need to work long hours to keep lesser jobs. The college degree may become merely one of the assets a brilliant kid may use to succeed.