In the US people tip waiters almost 20% while in Argentina this number is reduced to 10%. In other countries like Japan, tipping is insulting. And then in Spain, the customer might even be considered to be tipped: if you are with a big group and brought enough business, you will probably get some digestive "chupitos" (drink shots) for free at the end of your meal. Secondly, why are servers tipped, but not doctors nor salesmen? (Actually, in Japan doctors are tipped, but not servers...) These and many other questions were very well posed by Mr. Pink in Reservoir Dogs.
This begs the first question: why tip at all? History suggests that originally 16th century Europeans tipped in advance to obtain faster service. If you were in a hurry, you would put a few coins up front, making sure you are noticed so that you get better service. Some suggest there was a sign saying T.I.P., "To insure promptitude," which originated the word "tip". Others suggest it was actually a slang word that spread around. According to Michael Lynn, a professor at the Cornell University with over 50 academic papers on the topic, tipping in the US began in the late 1800’s, when wealthy Americans traveling abroad to Europe witnessed tipping and brought the aristocratic custom back with them to “show off.”
But nowadays the social norm is that we tip afterwards. Waiters are supposed to provide good service in advance with the hope of getting tip as a reward. Nevertheless, it seems that people tip almost automatically, a rather fixed percentage (which might depend on the country). A study by Cornell University found that quality of service did not correlate much with service. So it seems we do not tip for good service. One thing that did correlate with tips was how attractive the waitress was (not for waiters though). Touching the patron's shoulder when delivering the check also seems to increase tips. So beware of touchy good-looking waitresses next time you are in a restaurant.
Second, why do you tip servers but not dry-cleaners? Why tip the hotel doorman, but not the person behind the reception desk? Why tip a baggage handler at the airport, but not the flight attendant? There really seems to be no logical explanation for this. The U.S. is empirically tip-addicted, with 31 different services being tipped. On the other hand, Canada has around 26, Scandinavian countries between 5 and 10, Japan 4 and Iceland 0. Most of the world operates on the simple premise of a service charge or a fixed price, no tip expected. But not the US.
Being now in Japan, I can see that servers deliver food promptly even without a tip. The restaurant business does not run into chaos without tips. It seems that tipping is more of a social norm nowadays, rather unrelated to service, where some countries tip and some do not. Some services are tipped and others are not. But this social norm also seems to depend to the racial group of the customer. How much is enough for a tip? How much is too much? In the US, only about a third of blacks say they tip in 15-20% range, compared to two-thirds of white. This might be mediated by their socioeconomic status (lower average education and income), but it does not explain it completely.
Let me end with an open question I recently read. Suppose tipping had never been invented and you were starting a restaurant, would you use tipping as the way to compensate your best employees? Or all your employees? Would that be the system that you would pick in a vacuum to compensate your team? I guess not. But this rather odd system can become gigantic. For example, tips have been estimated to account for around for 40 billion dollars in the US, bigger than the GDP of almost half the countries in the world.
Deep question. And you might think completely out of reach of economics. But economics is fundamentally about how to allocate scarce resources in world of seemingly unlimited wants. Many would say life is invaluable. But health research, among others, forces us to think deeper. Suppose a drug extends life of cancer patients by a month on average but costs around 30 thousand dollars. Is it worth it? Assuming there are no other alternative drugs for simplicity, the question behind is how much is a month of life worth? This is not a hypothetical philosophical question. It was actually a case made public by doctors from the Memorial Sloan-Kettering Cancer Center.
If you think life is invaluable, you should then think that one month of life is worth those 30 thousand dollars. But think again. Resources are scarce. There isn't an infinite amount of money available. Suppose that money is coming from public funds, what if that money were used to help other people with simpler/cheaper health issues, which can also extend or improve life? What if that money were used in education? Now suppose that money is coming from your own pocket. Would you rather use it to go travel around the world instead? Or maybe buy a house for your children? Or, even, would you be willing to leave a 30 thousand dollar debt to your family for that extra month of life? If life is thought as invaluable, none of these comparisons can be done.
Doctors from the Sloan-Ketterin Cancer Center pondered about this due to a combination of new treatments available for cancer that were estimated to cost around 600 thousand dollars per year of life extended. Having been approved by the FDA, most insurers had to cover these treatments. But doctors at that hospital decided those drugs were not worth the price, other alternatives were better. If those new treatments were used by everyone, resources available for health would run out very fast. That money could be better used elsewhere. And so they decided to boycott them by going publicly against some of these treatments.
Avastin, $5,000/month; Zaltrap, $11,000/month; Yervoy, $39,000/month; Provenge, $93,000/course of treatment; Erbitux, $8,400/month; Gleevec, $92,000/year; Tasigna, $115,000/year; Sprycel, $123,000/year. (Photo: NYMag, Illustrations by Remie Geoffroi)
OK, so those treatments may not have been very good. Other alternatives were available. But what if the drug in question is really good and no alternatives are available? One such drug has been suggested to be SOVALDI for hepatitis C. Let me clarify that I know nothing about this drug, so let's approach the example more as a thinking process to understand how life sometimes needs to be given a dollar value, instead of a health study. SOVALDI is suggested to have smaller side effects and as much as 95% rate of cure in the US, hence an impressive drug. Its alternatives were suggested to perform much worse, possibly not curing Hepatitis C, just dealing with it temporarily. However, SOVALDI costs one thousand dollars a pill and is to be taken daily for 12 weeks, or around 84 thousand dollars per treatment. This has been taken by around 75 thousand people last year, totaling a cost of around 5 billion dollars in the US.
Given the costs, states have limited the coverage to some special cases. But the bigger picture is that this drug might actually cure you, reducing future costs and allowing patients to get back to their lives faster and with less problems in the future. So it might not be fare to just compare the price tags among the different Hepatitis C treatments. Assume away all other possible life improvements beside work and just suppose that the average patient is able to produce for one more year of life than those that take other cheaper drugs. Moreover, say this person produces the average GDP per Capita of the US: over 50 thousand dollars. This extra year of production can be considered to reduced the actual "cost" of this drug by more than half. Then add all other aspects of life that may improve with such a drug: not taking any more medicine later on in life, enjoying more time with family and friends, and so on.
The first drug was too expensive for what it provided. (Suppose it actually works) second might be worth the price. What is the cut-off? In other words, we return to the same question: what is a month of life worth? Neither zero, nor infinite. Many numbers are actually out there. For example, The World Health Organization typically places the value one year of life between one and three times the GDP per capita of the country, i.e. one to three times of what is produced by the average person in that country. I believe this number should depend on the person, determining the quality of that month (e.g. age, other possible health issues, general happiness): a month of life for an average 20 year old person should be worth more than an extra month of an average 90 year old one. But it gets very complicated to go into these details. However abstract this may seem, this number that values life is supposed to start being used to evaluate cardiology treatments. Independently of the final number, if you thought life was invaluable, think again.
Based on Radiolab's podcast.
A funnier example of scarcity, to finish off laughing...
Last week I wrote about the beauty of the most common measurement system in the world. In the Economics world there is also one measure that dominates all discussions: the Gross Domestic Product (GDP). Before the Great Depression this measure of "how the economy is doing" did not exist. People at the time felt things were going really bad, but no one could agree on how much. Back then they had numbers on how the production of some important industrial or agricultural sector was doing. Presidents Hoover and Roosevelt had to fight the Great Depression on the basis of sketchy data as stock price indices, freight car loadings or incomplete indices of industrial production. But there was no general number for the question "How bad are things?". Hence, similarly to the metric system, the base measure of "the Economy" was born during a crisis.
So the GDP was going to try to add up everything made in the country: houses constructed, beers sold, visits to doctors, etc. (In case you are picky, back then it was actually the Gross National Product, GNP, but I will avoid this difference here). And they needed someone willing to crunch all the reports and summarize them in one number. And so came an economist described by his best friends as "extremely dry": Simon Kuznets. He first needed to figure out what to count which is not as easy as it sounds since it should potentially include everything. For example, shall we include the Mafia? The US does not but some countries do. In 1987, after “black,” untaxed transactions were included, Italy showed a brilliant 18% growth and became bigger than the UK (il sorpasso ragazzi!). (I believe this was later undone) And the Italians might do it once again, given EU regulation from 2013 that requires all transactions, regardless of their legal status, to be accounted for. And another big jump forward for Italy might be ahead of us.
Another case of productive activity not usually included in GDP is unpaid work. As the IMF explains, for example a baker who produces a loaf of bread for a customer would contribute to GDP, but would not contribute to GDP if he baked the same loaf for his family (although the ingredients he purchased would be counted).
Well, suppose you have figured what to count out. Now, how do you count it? If it is produced and sold in the US but some of the inputs come from abroad you need to subtract that part (Tough job that requires some sort of input-output tables). Then you have all the goods produced in your country that you want to add up. If you were ever told you cannot add apples and oranges, Kuznets would tell you are wrong. That part was easy. You just need to transform them into a common unit, money, and add that up.
After 6 years of work (almost too late for the Crisis?), Kuznets published his best seller: National Income, 1929–35. Believe it or not, it sold out. And every country then joined the party and started putting numbers to the health of their economies. A first quantitative picture of the economies around the world was made. And the figure below shows how bad the situation was back then and how things have improved. Economists tend to look at GDP per person as a better measure of life standards (otherwise all big countries are going to be better than smaller ones). This chart shows the terrible effect of the Great Depression and the wars in Germany, USA and UK (PPP is to put everything in a common scale of prices).
Figure: GDP per capita (PPP, log scale)
Note: Scale gives the equivalent GDP value in 2012 US dollars.
Source: Maddison Tables and World Development Indicators.
But even if the GDP measure was too late for the Great Depression, it would come useful later on. The biggest test for this measure came in the 1940's with World War II. They needed to know how many planes and bullets could be made before prices would go up and goods in the cities would become scarce. The army thought they could make a lot. Roosevelt himself gave a speech in 1942 saying they would make 60 thousand planes and 45 thousand tanks. But Kuznets picked up his calculator and told the president it was not possible given the number of factories and steel mills. Of course the economist was right and Roosevelt backtracked. Did the confidence of the US on how to use its resources help win the war? Very likely. And all thanks to a very dry data-cruncher economist.
After this experience many economists thought that if they could measure it, they could control it. They thought it was like physics. If they understood which variables were moving the economy, they could manage them and drive us to the right place. After a century of several bad worldwide crisis experiences, it is probably safer to say if you can't measure it, you can't control it. But I guess that if they had been able to really control it, all us economists would be unemployed. Any mistakes done may just be for the well-being of me and my fellow colleagues.
(In the 17th century William Petty made some progress on some of the concepts behind the national accounts but his estimates did not allow to identify the magnitude of a crisis, the yearly growth nor the size of some particular product in a country's production, which are the main use of GDP estimations nowadays. Petty's objective was mainly to show that landlords did not make as much income as thought. He claimed labor was the source of 3/5 of national income and so if tax revenue was the objective, consumption and labor income taxes should be taken into account.
GDP is still a very messy object to calculate. Next week I will write on how GDP is actually calculated. Hang tight!)
How do you know how much a kilogram of coffee is? You probably use a scale. But how do you know all scales are the same? How do you know my scale and the scale used by Colombian sellers are the same? Fortunately there is a list of standard measures (including for example how long, heavy or hot something is) that keep all of us under the same standard. In this new post I will talk about the story behind the measures we all use nowadays. You may think I have gone way out of my field and you are probably right. But I just can't imagine any form of Economics without measurement. And I can't imagine measurement without thinking about the beautiful metric system.
Most definitions of these measures are actually quite complicated. For example, what is a second? The Bureau International des Poids et Measures defines is as:
Before then measures were based on Charlemagne’s ideas. Many were simply borrowed from human body, like the pied du roi (or king’s foot) or the toise (the distance across a man's outstretched arms). But what if men were bigger in one part of the world than in another? Hence measure were quite uncertain and clearly not fixed: they varied from town to town, between occupations as well on the type of object to be measured. So agreements on measures were hard.
What gave room to the Metric System we have today? An economic crisis of course. The famine of 1780s meant that food should get more expensive. But bakers were worried about increasing prices (lots of revolts were happening), so they started baking smaller loaves. People started noticing loaves were smaller, but no one could universally check their weight! And the French revolution set the reform environment which started with a new standard. They wanted a system based on nature, that avoided national vanities and could be used by all nations.
And so first came the meter: They took a quarter of the circumference of the Earth and divided it by 10 million. That's a meter. And this gave birth to the kilogram. To define the unit of mass they preferred water to other bodies (such as gold) because it was easy to get anywhere in the world. They divided the meter in 10, formed cubes of that size, filled them with water and voila! The kilogram. And from the kilo they defined other 4 base units...
This object in France makes sure that whenever I buy one kilo of bread from a shop, we can all agree how much that is. Well, unless you go to typical corner store where a kilo of bread may be less than a kilo. But even in that case we can actually determine the real weight and formally complain about it. This is supposedly the story of Poincare (also in France but in the 1900s). From Allen Downy's Think Stats book:
Let me finish with the story of the first kilogram. This "perfect" object has been used as a prototype to build a few other kilogram sub-prototypes (called sisters) over the world. And these have themselves being used to build others, all the way to our day to day scales. Every time each one of us checks his own weight, this can be traced back to this little object in Paris from the 18th century. And the most interesting thing is that recently it was found that (comparing it to its sisters) the perfect kilogram was losing weight! The funny thing is that even though it lost weight - since itself defines weight - the object is actually still one kilogram! Which brings us to the bummer conclusion that we have all gained weight in the meantime. As the definition of a kilogram got lighter, we all got heavier.
Charging taxes on income is hard. Worldwide experiences show that less developed countries have difficulties raising revenue from income taxes. Below I have plotted GDP per capita and Income (and capital) tax share of total government revenue for 2005. It is reasonable that most countries have a hard time making people and firms pay income taxes, but richer countries clearly tend to do it more.
Source: World Development Indicators.
The income tax has not been common throughout history. For example, a century ago the income tax almost did not exist in the US. Most government revenue was from trade tariffs and consumption taxes. These are much easier to collect. You just need someone at the port of trade, or some random controls at shops. Look at the sale value and take a share. But income tax is harder. You need people to be capable of keeping track of their income and sum it. And then you also need them to be honest and report it. Finally you need to enforce it, with a system potentially capable of checking every person. However, in spite of all these difficulties, income tax now account for over 55%. How did this happen?
Well, as usual, first came a government in financial trouble: wars are the starting point of most taxes. The idea first floated during the 1812 war with the UK, but it was unsuccessful. Later, the civil war was bad enough to ensure the introduction of the income tax (and the beloved IRS!), focusing on rich individuals. How did they enforce people to pay? By encouraging people to report their neighbors to the IRS if they were driving a Ferrari (or the horse equivalent of the time). Some people even claim that the income tax was key for the victory of the north. So the income tax might have even stopped slavery!
But this was not enough. The war ended and so did many of the pressing needs. You may think it would have been reasonable to keep the tax to build a safety fund? Wrong. As soon as the war ended, rich people didn't want to keep paying. Moreover, they could afford the lawyers and the Supreme Court agreed with them. Surprising, ha? But you can always count on new government deficits. They currency and stock market crisis of 1907 meant funds were needed. So just before the First World War the constitution was changed, allowing the government to collect income tax. But it was still focused just on the rich. Less than 2% of the people paid taxes.
Once the Second World War arrives to the American coast, more money is needed. So they decided to expand the income tax to the middle class. They needed someone beloved, with credibility, charming to promote the tax: Donald the Duck. Yes, you read correctly. Here is the video:
And this is how the income tax came to be in the US. With an approval rate usually above 80%, it surpasses any politician I know. As Walt Disney appropriately said: "If you can dream it, you can do it."
Do people get paid more if they are better looking? Freakonomics did a podcast interviewing beauty economist Hamermesh who explained that:
If you are an economist you may think the statistics above could be problematic. For example, better looking people could be more charming or better communicators, which are more "standard" sources of income differences. However, even when looking at quarterbacks - where looks may not matter as much, except for merchandise/publicity revenue which is not included here - Berri finds that beauty (based on symmetrical faces) pays as well.
So, if surgeries cost less than $10k, and we can increase our lifetime income by over $150k, shouldn't we all get a surgery? It seems like a great investment. Well, data (from China and the surgery mecca of "North Korea") says this does not help...
Planet Money did an excellent episode on how much university majors pay. Some great data:
Follow me on twitter at @diedaruich
News and posts for an active mind.