Real wages, politics, and health: the McWage project
Real wages, politics, and health: the McWage project
Real wage rates show what workers are paid in units of consumption goods. Using the wage rate of a McDonald's restaurant worker permits a comparison of real pay for identical jobs across 70 countries. A comparison across US counties is a powerful predictor of health status and political behavior.
good evening everyone and thank you for uh attending this event at the festival of economy and trento we have our guest today or leash inventor who is one of the most prominent economists from princeton university new jersey one of the top american universities and he's specialized in a number of economic areas and mainly relevant to law labor and relating subjects he was the director of the american economic review and a member of a number of economic associations and again is one of the most prominent american economist but if i may add a personal note what makes him very very nice is that he's an expert in wine he actually created an indexer to evaluate the quality of the french wine using econometric parameter considering the dryness of the the the air and you know the humidity of the soil and so so forth it's currently the chairman of the american wine association um so it's not just entrenched in the other tower of economists but is part of the real life today we'll be talking about a very interesting topic that is to say the evaluation of a purchasing power he carried out a study based on the wages paid to mcdonald's crew mcdonnell as you know he's a multinational group par excellence operates in more than 70 countries i believe the 70 countries are the countries that were taking into consideration the study and on these grounds professor ashenfelter considered first of all the rate the wage rate paid to people around the world by mcdonald and the impact on national health systems well in a number of countries as you know is no national system people have to provide for their own health insurance mcdonald is a used to be involved in international indexes you might be familiar with the big mac index evaluated by the economist every year and this is something different it's the cost of big mac it's also called the burganomics actually so it's the cost of big mac which is the same around the world though it has been uh priced differently in the various countries and on the grounds of the mcdonald price um the purchasing power of a certain currency calculated of course not officially i happen to read in the economist recently that the latest index is the following 5.6 dollars in the us at 2.83 dollars in china which means that that the yuan is underevaluated by 44 at the current exchange rate so what you may say well this is an interesting way to understand the real purchasing power of a currency based on a tradable good that is a burger in this specific case using the different currencies there but going back to professor ashley felter he will be talking about the real wages this has nothing to do with big mac index of course but professor ashfield has evaluated the wages paid to mcdonald's accrue and has drawn a number of interesting conclusions that they will be hearing now i can tell you that we decided to have a first contri the contribution by the professor and a short q a session to take also the questions from the floor and then we'll be resuming with the second part of the presentation and we will draw in conclusions at the end so i have the prerogative of the speaker uh there seem to be two chairs here that are not being used uh if if the uh if it's permissible perhaps someone wants to take the i rather sit in a chair myself uh maybe you can take that chair if you like it someone who's no this person is standing back here there's a seat there i know it's reserved but no one's here so i would say until they come there you go perfect the best best seat in the house uh i hope i don't make you hungry uh i know i'm glad you already had lunch because i might make you hungry i i may i put this deliberately that that's called a full meal in some languages uh i put that uh it's very small so you would not be tempted to worry about it um let me start uh just by saying uh why we want to study a real wage a real wage is the amount of consumption good that you can obtain by working for an hour so if you want to think about it as an algebraic matter it's i'm going to use dollars uh it's the it's the dollars per hour divided by the dollars of some good which means when you divide you get the number of goods that you earn per hour so in that way it's if the goods can be made comparable in all places at all times we can make comparisons over all times and across all places but in order for that to be true we need the work the person does to be exactly the same and we need the good to be exactly the same that's how i came as i will show you to think that this giant uh conglomerate mcdonald's which operates it receives much more of its revenue outside the united states than in by standardizing things provided a measurement tool for us why do we want to know this well first of all it allows us to compare and i'll show you how living standards over time and across places in principle it it provides a universal measure it also allows us to measure the effects the effects of economic changes reforms in some countries maybe setbacks in other countries so it gives us an independent way to measure that now that's not to say that the price of a big neck cannot be manipulated by a government that desires to do that but so long as it doesn't happen we have a independent valuation that's how i came to think about this finally one of the most interesting things about the wage rate is that it's also a measure of productivity uh generally speaking uh the more productive a country is for people doing exactly the same thing the higher the wage will be so if someone who works at a mcdonald's makes more money in italy than they do in china which i can assure you they do that must be because italy is a more productive economy there's no other difference between them that's a study of total factor productivity and the real wage is one way i'll show you more about this uh is one way to get at that at that measurement now let me give you a small outline of the talk uh a little history about real wages the study of real wages goes back a long way and the reason is because it's one of the few economic measurements that we can use that has been used for for hundreds of thousands of years for example diocletian the roman emperor provided some of the earliest information we have on wages and prices because he engaged in something that richard nixon also engaged in which was wage and price controls in order to establish wage and price controls he had to set up what would be the maximum wage and what would be the maximum price for many different products so we actually have measures that go back there's a very lively debate right now about what the real wage was uh in the roman empire so you can study these things much more with much more care far more convincingly than you can think about gross national product in 500 doesn't make much sense secondly i'm going to do a little bit of economic theory so there's a little bit of heavy lifting uh you need to concentrate a little bit at the beginning so you'll understand exactly what i'm trying to measure and how you should interpret it then i'm going to show you results in the usa and i have actually a new project brand new data that i will show you has never been revealed before which provides information across the whole united states we've actually collected data on the wage rate at mcdonald's in over 13 stores 13 000 stores so i have many thousands of observations in the u.s then we'll break uh and at that point we can have questions and then i'll i'll turn to uh an analysis of the rest of the world and some other things you might want to do with this kind of with this with the set of data let me start with a with a with an example which was given to me by a man named bob allen it's a little bit hard to read so i'm going to help you if you have good eyes you can probably see it this is a comparison in 1704 of canton and london and it demonstrates two points one it demonstrates that the construction of a price index is complex because it depends on what it is that people consume in the two places that you're going to compare so in the first column there are for many things that would have been considered necessities the ratio of the price in england to the price in china canton is about where hong kong is now so this is like london versus hong kong uh starch by the way has to be different because starch is bred in england but it's uh rice in china and then there's meat milk and so on as you go down the list you'll see this is the price in england compared to china prices are much higher in incidentally they're measured in pieces of silver that was the coin of the of the world in 1700 little pieces of silver so they're measured in pieces of silver and you'll see at the bottom but it looks it looks like the english prices are higher all the ratios with a few exceptions like milk or milk is relatively cheap in england compared to china everything else except for charcoal is more expensive in england at the bottom i show you the relative wage rate of an ordinary common laborer you'll see that the english wage rate is about 3.6 times higher 3.6 the ratio of the wage rate in china so the english workers made more money but you can see right away they also paid much higher prices for the things they bought now in order to construct a comparison we need to have a weight how much did they spend on each of these things so what's in the middle column is the english budget share how much was spent on each of these products and in the third column is the chinese budget share it'll be no surprise to most economists the things that are relatively expensive in england are things that they consume less of so starch is much more expensive relatively uh than other goods in england and you can see the budget share in china on starch is higher than it is in england milk is exactly the reverse chinese drank virtually no milk english consumed a lot of milk this is an inevitable product of the fact that the relative prices are different so which budget share should i use if i use the english budget shares i get a price index at the bottom of three consumer price index the english prices are three times as high as chinese prices and if i divide that into the wage to get the real wage i get 1.2 something about 20 percent higher wage in england than it is in china but if i use the chinese budget shares or the chinese have now optimized themselves the price and relative prices in england are nearly are more than four times as high almost five times as high as they are in china and if i divide that same relative wage i find out that the chinese actually the english actually makes 75 as much as the chinese now a normal way to deal with this this is an extreme example but is one i want you to be aware of a normal way to deal with those would be to take an average an average says they made about the same real wage in hong kong and in london virtually the same in 1700 in fact i'll show you generally speaking real wages were the same everywhere in 1700. there were no differences across the world most people lived for those of you who remember uh ricardo and malthus they wrote a wrote a theory basically of how it was that wages would always be a subsistence well they were not ignorant they knew that wages had always been at subsistence the theory came from the facts not the other way around by the time that they wrote their theory down of course it was wrong typical of economics you should make your theories in advance not not after you know the facts in any event uh let me show you one more little piece of history and i'll finish this this is one of my favorite examples because it involves two famous labor economists one was my teacher albert reese the other one was paul douglas paul if anybody is an economist you've probably heard of the cobb douglas production function that's mr douglas douglas a remarkable man was a professor at the university of chicago and during the second world war he enlisted and served time uh in the japanese theater he came back and became a senator from the from the city of from the state of illinois so he ended up as a as a as a senator he had constructed from the period 1890 to 1914 douglas had used his own money to construct an index of real wages in the united states 1890 to 1914 is a quite important period in economic history 1914 as the origin of the first great war we the americans did not come into that until later but many economic historians treat that as sort of the end of normal times 1890 to 1914 is a period of what we would hope would have been industrialization and rapid growth in uh the u.s now douglas and reece douglas started and made up his index reese came along that's the blue line and you can see nominal wages which are very low by the way about maybe 20 cents an hour very very low no but going up nominal wages increasing uh reece decided that there was something wrong with what douglas had done and let me show you why he thought that the the douglas's green line is an index of prices and douglas's index of prices as you can see is also increasing dramatically so that if you look at the red line in this graph that's the real wage rate as measured by douglas it's flat there is no progress in the united states according to douglas between 1890 and 1914. reece did not believe that could be true so he went back and reconstructed all the work that douglas had done adding new sources the new source that raised the wage you can see it doesn't change the relationship was using not just union contracts but also other measures of wage rates and that didn't change things much but what really changed things was prices douglas has prices increasing the green line very rapidly in this period reese does not what's the difference well reese hit upon what i consider to be in retrospect a brilliant idea he used as a way to collect data on prices of identical items the internet of 1890. now you did not know there was an internet in 1890 there was exactly like amazon it was called sears roebuck it was called montgomery ward and what they were is gigantic catalogs thick like this that came to every house in those catalogs everything you would want to buy is there you could buy a house i remember them i'm old enough i remember them when i was a child they sent these catalogs just before christmas smart just a little before christmas and i looked at the toys toys toys toys toys begged my parents for the toys uh you know the little uh crystal radio i don't know if you've ever seen a crystal radio i begged for the crystal radio i did get the crystal radio i was very happy uh thinking that in the middle of this i lived in north dakota at that time which is in the middle of nowhere i could collect get a radio signal of course i couldn't uh but i tried my best so he used those catalogs over the period of 1890 to 1914 to create a price index where you could use the same items year after year and he could take account of the fact unlike douglas who had used mainly commodity prices that there was a huge introduction of manufactured goods which in fact did not increase very much in price just as we've seen today for many many things that are easy to manufacture have become incredibly inexpensive i just want to make one last point so reese's real wage index is the blue line here and you can see the real wage grows so the amendment is yes during industrialization real wages were growing in the u.s probably as they were many places the other thing i want you to look at is the yellow line the yellow line this is also something most young people don't know historically when people were working at subsistence which people most people did throughout throughout all of history it was normal the work hours were sun up to sundown every day but sunday that was the norm and even in 1890 the normal work hours were 60 per week you work six days a week for 10 hours you can see this yellow line is declining over this period 1890 1914 down to 55 hours per week basically means they went over more or less to half day of work on saturday one of the sure signs that a country's real wage is growing is that hours at work are declining leisure is a superior good it's a good that you want more of and if you get richer uh it will be the case that you'll take more time in leisure so people today think they work so hard they don't work so hard nobody's working 60 hours well maybe a few graduate students here at trento but otherwise no i don't think so let me just finish with this story about when the takeoff took place this is a a a composite road wage rate measure uh for many places japan canton beijing delhi and so on florence italian one and it shows you the percent by which the real wage exceeded subsistence what there's a measure of this basic bare bones subsistence and you can see in 1914 most countries were still quite poor including italy the countries that were rich were the english the dutch and the americans in this chart japanese were not the japanese economic miracle took place after the second world war when they entered the second world war they had no chance against the firepower of a country as large and as well off as the u.s so now what you can see many of these countries things have changed japan is now the more or less the same as all the developed countries italy is more or less the same as all developed countries uh beijing is not yet there but it's growing quite a ways below uh kenton hong kong is much higher mexico city is still has not grown very much colombia has not grown very much so what we've observed is some countries have managed to grow many have not and i think the long-term goal for economic civilization is that everyone should be at that level of development which i'll show you in a few minutes equals two big macs per hour now let me take a little second to do a little heavy lifting theory there are two ways to think about a real wage measure one is to think of it as what what how much do i need today to be as well off as i would be in some other country or in some base period so what i think of is some w star that tells me because prices may have changed how well off i need to be today uh how much wedge i need today to be as well off as i as i as some other country or in some other time and then what i do is i compare the actual wage rate to that wage rate you need in order to be well off a typical if it's the only thing that changes is prices then you can compare wages to some price index notice this interpretation of the real wage rate is not is not affected by uh the failure of competition in the product market or the failure of competition in the labor market or the existence of minimum wages if there are minimum wages and they raise the wage then yes the real wage went up this is a measure really of worker welfare but there's a second measure second way to think about the rule oh i want to talk about one other thing this is the most important point take away that most people don't know uh tradable and non-tradable goods this is i'm gonna i'm gonna try to describe to you this bolasis samuelson theorem the basic idea is this some goods like the iphones you all have are tradable some goods uh like uh the meal you had at lunch parts of it are tradable but parts are not a haircut everybody here has i need a haircut by the way if you could find a barber uh haircuts are not tradable a big mac is an interesting example of something it's it's and i'm going to use it as an example because i'm going to use it a big mac is a quasi-tradeable much of what goes into a big mac is tradable the bond the meat so on the equipment it's tradable there's one part of the big mac that is not tradable and that's the worker so there's two components to the big mac one is local and it's produced by the local worker and the other is tradable now when we think about comparing people in different places we need to think about comparing them not using iphones so if we measure wages and iphones the chinese are abjectly poor but if you think about measuring real wages in haircuts they're very rich what we really want is something in between and that's what the big mac is it's a quasi-tradeable part of it is tradable and part of it is not as i'll show you a big mac is about half the price in china as it is here this has nothing to do with exchange rates the tradable goods which is not what a big mac is the tradable goods prices are more or less the same the only thing that exchange rates are designed to do is to make tradable goods the same price they don't make non-tradable goods the same price so the belossus samuelson or pinworld trade uh 10 pin world table effect is one that implies that the real exchange rate will grow as a country has developed the basic concept is countries differ in the productivity of the production of tradable goods some countries are good at producing tradable goods and and others are not poor countries generally are the ones that don't they're not efficient at producing tradable goods though you think of norway or sweden efficient at tradable goods europe mostly efficient at tradable goods india china not so what happens is as the efficiency in the tradable goods sector grows the country will grow but wage rates have to be the same in both sectors so the productivity of people in the tradable goods sector determine wage rates in both the tradable goods and the non-tradable goods sector so it's very important to think about it because when i talk to you about the wage rate at mcdonald's measured in dollars which i can do in all places because of exchange rates that tells you how much that wage rate is measured and tradable goods but if i then ask you well let's measure it in big macs that tells you how much workers how well off they are in in a quasi-tradeable in sort of in a purchasing power sense these are two completely different notions of what we want to think about as a real wage one one ex it's about productivity and the other one is about how well off the workers are okay i think that's enough uh oh let me say this is basically i told you this uh the usual way of thinking about it is the role wage is proportional to marginal product of labor and if you think of total total factor productivity is what makes countries the free lunch it's the thing that makes the country um more productive but doesn't require any more inputs it's about productive efficiency in a much more general sense than just a measure of capital it's it's uh it's efficiency in a far more general sense and we know countries differ in those ways because we know that wage rates differ for workers that do exactly the same thing and they shouldn't uh if if they were not total factor productivity differences um so that's the way i'm going to think about the real wage the wage measured in dollars i'm going to think of that as an index of total factor productivity for a country and i'm going to show you by the way that the mcdonald's wage matches many other measures of total factor productivity someone is a testable proposition it wouldn't have to be true but it is true independent measures of total factor productivity look just like the wage rate at mcdonald's and then i'm going to show you some measures of how well off people are in which i'm going to be using big macs per hour and i'll show you some other measures of other wage rates to show you that there is really a wage rate in almost every country um oh well there's the product oh it looks big now doesn't it i'm going too fast do that again i'll try to be slower um so there's the product um this is the first place where i collected data on mcdonald's do you know where it is no it's india uh the ma the big mac is not made from this animal you cannot eat these animals uh the uh but i went to that mcdonald's um and i bought a big mac and the reason i went was this i was doing some consulting work pro bono consulting work with mckinsey on productivity in india and indians love to argue about inequality they love to talk about poverty so i was at some point i was just could hardly stand it and i said to a former student of mine who who named sergey bala an indian i i asked him well look do you have mcdonald's yes we have mcdonald's okay how much does someone who works at a mcdonald's make i don't know let's go find out so we went out to this store there is a picture these are the full meals by the way so this is the you see the can you see the uh the chicken maharaja mac the the the the big mac is called the maharaja mac in in india perfectly sensible and it's made from chicken actually when i went there this is a little bit later period they made it out of lamb which was really good i thought it was one of the best big macs i ever had you could probably tell from looking at me i have eaten big macs this is the price of a full meal and that's my cash register receipt i paid for this meal and i got that receipt that was in 2000. you can see it's right on there july of 2000 and you could see my well my my indian friend is a vegetarian so he had a vegetarian combo thing some something i don't know what it was i had a a maharaja mac and a shake which is my favorite thing uh and then he had a coke and you can see i paid 52 rupees for my uh maharaja max you know how many how much how much what i kind of know what that would be in euros but in dollars rupees are about 40 to the dollar so that's about a dollar or 50 for a big mac i will bet you that over here in trento uh where you can get up there's a mcdonald's here i know uh over here in trento i'll bet you pay 450. so the indian big mac is a very good deal however there is a problem you cannot get the indian big mac here i would recommend it here but you can't get it so that's an example of a quasi-tradeable it gives you the idea of that i show you this because uh i that's how i got interested in collecting data this is my favorite mcdonald's in the world does anybody know where this is it's in prague that's a listed building and this is the man uh czech man stefan eurada who i've done much of this work with is standing in front of our favorite mcdonald's i have to go back like a pilgrimage i have to go back every time and have a big mac and see how it looks uh why why make wages well first of all the interesting thing about it is we use the entry level basic crew pay skill inputs are identical people there's no change in the technology used to produce mcdonald's the hedonic job qualities are the same you can't basically do the same job i've been in them in india and china u.s everywhere things are more or less the same um the uh and there is a reason for this there are over 140 countries i've collected data in about 70. there there's a reason why they do this the same way i don't know if you know what it is the origin of mcdonald's and its growth is food safety food safety the thing about mcdonald's is that because their reputation is worldwide they cannot permit anyone to violate food safety rules so the reason that you might want to eat mcdonald's in india is because yes you can drink the water yes you can eat the food you will not get sick i'm here to tell you that the same is true in china uh when we arrived here someone asked in where we're staying and being airbnb can you drink the water sensible question in half the world you cannot drink the water you can drink the water in all mcdonald's that means by the way that they have to filter all the water in poor countries and they have to have generators because most of the stuff that they use is frozen they this is all run through what's called their operations and training manual super secret uh where when they break into a new country they take all the people the ones that are going to be local and they go to hamburger university hamburger university is outside of chicago uh it uses 20 different languages in which people are taught how to do exactly what you're going to do all over the world they're growing now very rapidly especially in china in fact at this point there are more mcdonald's in china than there are in the u.s just to give you the scale of what this is like but now remember china of course is five times as big as the united states so there may eventually be five times as many um they don't adjust technology generally and they and by the way the the handbook tells you to not violate any local laws but it does not tell you what to charge for a big mac and it does not tell you what wage rate to pay there are no rules about that local you do whatever you want how's the data collection work we've collected data in 2007 for 64 to 66 countries it depends iceland dropped out of our sample during the financial crisis they couldn't do they couldn't afford to import the materials to make big macs and they all went out of business uh venezuela has just gone out of business also there are no big macs there used to be big macs in venezuela they're now gone uh we we've collected 27 countries since 2000. so i have some data to show you over a long period and then i have more data to show you over about a 10-year period we collect the data hourly wage of the crew member price of a big mac we collect data from the urban areas in most countries and we concentrate on the the correlation the median and the average is about 0.99 we have regional data in some countries now we have it all over the us but we have been collecting regional data in india china for a long time we've also collected data from starbucks i don't know if you have starbucks i couldn't find one i really would like us you know if there's a starbucks in trento please email me i don't need the big mac but i could really use a venti coffee the the uh and that's to show you and i'll show you that to show you that what's going on in different countries is a wage rate in other words it's not like there's something special about mcdonald's uh we've done a lot of work on liability and here's where i'm going to make you an offer that'll be difficult for you to refuse i do not have the wage rate or the price of a big mac for trento if you go to the big mac to the mcdonald's in toronto i'm only going to do this for five people and you're one of the first five you go to the toronto mcdonald's you have a big mac anything else you want big mac you send me the bill and i will reimburse you for your meal you can find me online i can see some people are going to leave the someone may leave right now don't leave yet lock the doors uh you have to do one more thing you have to ask a crew member what the wage rate is and then you write that on the back give me the address of the store and the and the date will be on the cash register receipt you send it to me you can find me very easily google me early action felter it'll be in princeton and for the first five that i get i will send you a reimbursement either a mail order or maybe even cash i don't know whatever it is i'm going to guess it'll cost me about well now that i've you can have anything you want i would recommend a shake fries and a big mac that's probably going to cost me about ten dollars it's okay i can i can afford that and then i'll have one more data point because we only in italy we only collect data from milan and and rome uh okay now let's just remind ourselves of the wage rates we're going to have a wage rate measured in a common currency i'm going to use dollars uh there's going to be a price of a big mac i can put that into uh dollars that's going to be a the measure of a non-tradable and then i'm going to also calculate so i'm going to call the wage rate in dollars and go that much wage the big max per hour is going to be the division of the wage rate in dollars by the price of a big mac in dollars so it's going to be the number of big macs that you get for an hour of work the couple limitations uh there are minimum wages in many countries like denmark or france if they're not real minimum wages their industry standards it's important to understand that from the point of view of the welfare of the worker that still tells us how well off they are but it does cause a problem for measure for the measurement of productivity because then the real wage is distorted off of the off of what the productivity level would be measured from the production function uh secondly the fast food price there are occasionally signs that fast food prices are being manipulated that does occur argentina is a famous example because the economist collects data on big mac prices it's very tempting to control the price of a big mac and make your exchange rate look like it's different from what it really is so argentina apparently the big mac price has in the past been controlled they have to offer you a big mac but uh they don't allow you to find out what the price is they stick it like around the corner somewhere and discourage you from consuming it so that that also is a imperfection in the in using it that way let me just show you a couple of things about reliability this is the muck wage this is dollar wage uh against starbucks wages you can see the countries are labeled here uh the poor countries are at the bottom uh china has starbucks by the way quite nice uh very good wi-fi i suppose the you know the communist party is looking at everything you do but i wouldn't give them your credit card but other than that you can certainly send love to your friends um you can see that the wage rates line up perfectly so people at starbucks make more money than people at mcdonald's generally speaking not always but you can see they line up very nicely so if the wage rate at mcdonald's is higher in one country the wage rate at starbucks will be two so there is this is this is i think convincing for almost everybody that there is a wage rate in the country there is a sort of a wage rate for that minimum skill worker called that this is now this is a little different this is big max per hour versus coffee per hour you could think of it as the big macs per hour versus a venti that's what i use per hour it's not quite as close uh the uh for example there's there's some extraordinarily high coffee per hour prices in the us and the uk but most of the others lie right along the line uh here's an example of manufacturing wages uh if you look at the at the at the uh this part right now isn't working i don't know well the lower right part those are the high minimum wage countries so they they have wage rates that are very high compared to manufacturing and that's because they have minimum wages in the manufacturing wage is the one that's set by the market these are the icp this is the international comparison project i do a lot of work with them now uh this is where the the poverty rate for the whole world comes from the international comparison project they they develop purchasing power parity indices in fact they're going to use some of our data to do purchasing power parity measures inside the united states and you can see they line up pretty well uh that that one is actually i think it's for the component of the icp which is cleaners this is for cooks so you can see even in the poor countries which they cover well they're related now let me show you a little bit about the us and then we'll take a break what i'm going to show you next is quite extraordinary recently collected data and it surprised me let me start with a map of our great nation uh the darker the color the higher the big max per hour you can see the picture this thing here oh i turned it off oh wait it must be this one yeah there's the red dot okay so you can you can see the picture the band of real wages is to the west the north and the east the middle is lower and i was surprised these big mac per hour differences are quite large the highest is 2.4 the lowest is 1.6 mississippi is the lowest uh i think california is the highest these are these are differences 2.4 to 1.6 these are differences on the order of uh uh uh almost uh three quarters of a big mac there are very substantial fractions so we have and so i think the way to think about this is new new estimates the way to think about it is the us is just like the rest of the world some places are are more productive in tradable goods manufacturing they're more better at tradable goods production so if you think of california's movies tradable good computers tradable goods software tradable goods you think of mississippi well you probably can't think of anything they're not effective at any kind of tradable goods production you think of minnesota that's the picture up there by the way the high wage one fire it's all kind all almost like uh if you do if you get a valve replacement you're going to get a saint jude's valve it's all made there a male clinic is there many manufacturers of medical equipment are located in minnesota very efficient place i think the way to think about it is we we just like the difference is not as stark which has between china and the us basically some of these places are less productive than others and because now here's the part that's surprising is because people are not mobile across state they don't move they should move but because they don't move real wage differences can persist over quite long periods uh and as much as 1.6 to 2.4 is pretty remarkable uh we can measure the wage weight really beautifully here we've done the survey we do for example we take 500 stores and go back and do it again see how highly correlated they are do a great job of measurement now so this gives you a general picture now i should add one more thing where you see what appear to be anomalies like this that state up there that's what that's west virginia now west virginia is a very poor state why why do they have such a high wage and the answer is they passed a minimum wage many of the states that have another good example is nebraska and that's this place here they actually passed a constitutional amendment minimum wages are extremely popular with americans but not with american politicians so a typical example is the state i live in new jersey where the legislature passed a minimum wage increase the governor who was republican chris christie wanted to run for president show his republican credentials vetoed the law they did not have two-thirds of the majority in both houses so they put it up for a constitutional amendment the minimum wage in new jersey is in the new jersey constitution because the people voted for it the legislators or the governor did not and that happens in many parts of the us sometimes there's a city minimum wages are very popular and nebraska for example is a 100 percent red state there's probably not five democrats in that state and they did exactly the same thing their legislature would not pass it because as a matter of ideology minimum wages are bad people wanted them so the people took it in their own hands and voted either a change in the usually they have an initiative so some of these are exceptions california typically is not and they do have a minimum wage but most wages are higher than that now let me show you something more remarkable this is the this exact same data set except now it's broken down by county so these are 8 000 counties in the united states if there's no no color that means there's no mcdonald's so in the vast center there are places especially in the dakotas uh up here the white that you see that there's nothing there there are also no people there so not much going on uh so i can't measure mcdonald's there but you you can see this shows you more or less the same picture uh real wages are high along the west across the north and then down into the east and a little bit higher in florida is that florida there we are can i find it i can't find the red value a little bit higher in florida than the rest of the south now let me show you an amazing thing too i'm going to show you two amazing things i'm going to show you another map the next map is mortality rates at age adjusted mortality rates by counting remember blue here means high okay high wage rates the mortality won't be the same way blue means high but you'll see they're inversely related you know the wage you know whether you're going to die it's a remarkable finding uh there are little isolated spots but generally speaking the places i showed you the low wages in here this is the heart south and that's the places with very high mortality the low mortality is exactly the place where those real wages are high now the sec that i want to show you one more map and then we'll break the second this map i want to show you now is to me to my mind it's even more remarkable it's the fraction of the vote that went for the republican party by county you can see it looks almost the same there's some exceptions up here there are not very many people there uh wyoming utah and so on but generally speaking this area right here very poor area certainly not california not too much up here and not much on the east it's quite remarkable because it's almost the opposite of what you think if you watch what donald trump has been doing it's clearly not doing much to assist this incidentally uh that you may not have heard this we can ask about it in questions uh someone can ask me if you'd like what what does donald trump think about the minimum wage and i think that we can break at this point oh i wanted to show you there this is just two things uh we'll leave it right there so let's go back we'll come back and then i'll show you some i'll show you what the what the wages in the in chile and and the rest of the world and what's happened to growth take a break so you stopped at the nice at the most interesting point to namely when you asked what does mr trump think about the minimum wage well the what you showed us impressed us namely there is a correlation between high mortality rate and the low wages are there any questions from the floor is there anyone wishing to ask a question we have a prominent representative of the italian uh one of the italian major trade unions here with us maybe she would like to ask a question anyway short questions because we don't have just out of curiosity why in some graphs the netherlands was sticking out to the point thank you okay okay uh that the the the graph i showed you the only place for the netherlands show amsterdam showed up was in the data for 1914. uh i do have data for holland but it'll look more or less the same as the rest of the western european countries so that example was not anything about current information oh so why why do you think of your questions i would like to ask a question to professor ashley you showed a diagram or a graph relating to the productivity increase and the reduction of working as uh in the second industrial revolution so between 1890 and 1914 considering that the first industrial revolution took place a century earlier currently we are undertaking a new industrial revolution based on the labor automation and an alarm has been sent about the loss of jobs that will follow as a consequence thanks to internet of things and big data artificial intelligence uh self-driving cars and so on and so forth so there has been a substantial and remarkable speed up in the automation that would steal away jobs or destroy jobs as was the case years ago centuries ago when the industrial revolution actually destroyed jobs in the agriculture so do you think it would be possible to recover the jobs that will be lost thanks to new jobs that will be created by the new technologies i know it's a big general question but maybe we can you can provide us with some food for thought um it is a big question very big question uh well it's but it opens up this discussion which has been going on about um whether we're in the some people think we're in the beginning the potential beginning of a great productivity growth uh and some people don't you know the the favorite comment uh is uh you know in the 19th century we got the steam engine and in in the in the 21st century we got facebook uh is this facebook is facebook create the productivity of a steam engine probably not so that's bob gordon his view is that productivity growth is slow i think many economists have changed their views more toward gordon because productivity growth measured productivity growth is very slow it's not very it's not substantial in any country so although there is concern about robotics i think it's concerned that if somehow is not increasing productivity i think historically so two things to say about that first the productivity growth is not very high secondly the link this will be interesting perhaps for the trade unionist the link at least in the us between productivity growth and wage growth which when i was a student was taken for granted uh no longer seems to show up in the data so we now see productivity growth when we do see it and we still don't see wage growth i will show you in a little bit data on wage rate at mcdonald's but uh the last decade has not been good news and i don't know the answer it it's it's an open question whether uh we can find something that will generate productivity growth and do it rapidly enough compared to population growth so that we can actually see increases in real incomes now in the u.s the the the breaking of the link between productivity growth and wages has meant is identical to really uh the decline in labor's share of total income so we have seen a decline in labor sure of total income i cannot believe it will continue to decline in which case if there's productivity growth then there has to be wage growth but who's to say okay prego thanks a lot very very fascinating in fact you cut in at the most interesting spot basically my question is is there more than a correlation so is there a real causal link between higher wages and higher mortality because let's say for my poor understanding of the united states those are all the most liberal states where there are also maybe more public goods like california massachusetts and so on so is more than just pure link uh israel do you really find a causal one that's a very good question um so many people have been looking at the correlation between mortality and other things most measures of social failure are correlated with mortality measures uh i think this is a particularly interesting one because it's not identical in any way to mortality it it's it's a and if you interpret the way i did which is why would there be real wage differences across american counties well it must be because that county is more productive in some sense and being more productive relies on all those things you mentioned it relies on infrastructure everything else so it may be that the real wage at this level is a measure of total factor productivity at the county level something you could normally never get and it may be that that total factor productivity is what's driving the mortality and then and that as you say the fact if you just raise the wage it wouldn't by itself do anything uh now one way there we haven't done this but one way to test it is to is to look at whether there are two ways really one way would be to look at the correlation where the minimum wage is generally not binding and see whether it's still true that the wage rate is correlated with mortality another way to look at it would be to look at the change in the wage uh and compare it with any changes you see in mortality now it could be the mortality change takes a while and you should look at long wage changes that's harder to do but we will see minimum wage increases in many u.s states when we collect we collected these data in 2016 will now collect again in 2017 and there will have been many minimum wage increases over that period it's unlikely it would show up in mortality right away but if we collect data long enough we might be able to look at the changes which i think would be a more convincing uh exercise the way i interpret it now i think if i if i if you want to have a model in your head just think of a model in which the productivity in producing tradable goods differs across counties and it's that productivity which is related to everything associated with people living in that county uh that would make the mortality lower so the causal direction is not all together clear but i think it's it's probably to at this point it might be the most independent thing we have that we can correlate with mortality at quite high correlation by the way across counties the r square is like point three the uh because everything else we have is you know like suicide rates or something but those are more or less the same thing as the mortality that you're looking at but that's a very good question it's not clear even if it is there's a causal link it's not clear what the interpretation is all right let's listen to the last part of the presentation thank you very much so let's listen to the last part of the presentation thank you very much big mac index in the meantime the big mac index i got here shows that india is 44 cheaper than the united states means the rupee is a 44 less value than the united states currency you missed my point it will be until productivity grows um okay let me show you a few numbers uh in other words the point i was trying to make earlier is that the price of the big mac reflects two things tradable goods they have to be the same everywhere and non-tradable goods which is labor if labor is cheap you can do the algebra if the if the wage rate in india is uh it's probably 30 cents an hour 40 cents and if a third of the cost of a big mac is labor then there's going to be a massive difference in price just just because the domestic good the domestic the untradeable uh factor is so inexpensive so let me show a few examples this is current uh 2016 for the usa the average wage is 10.50 uh the price of a big mac is on average 4.85 so this is kind of a benchmark 2.16 big max per hour give you some idea generally speaking is the way to think about it uh big max per hour is about two in developed countries so just to show you that that's right there's italy the wage rate is lower 9.8 but the big mac price is lower too and so you have big mac per hour of 2.12 a little bit lower two percent lower say than in the us very similar now let me show you some other countries that are maybe more interesting there's china the price the wage rate for a person who works at mcdonald's is two dollars and twenty cents an hour but the big mac price is half the price almost half the price of what it is in the u.s or china so big max per hour is 0.84 nowhere near as high as the u.s or italy uh maybe uh a third uh maybe a little bit a little bit less than a third of what it would be and china is actually doing very well as developing countries go here's brazil many people think of it as the rich country it's actually a very poor country uh wage rate is a dollar 31 an hour big macs are very expensive and big mac per hour is about 0.27 very low uh i mean even if you didn't adjust for the big mac price you can see the the wage rate in dollars is almost half what it is in china that's that's by the way dramatic change over the last 15 years there's chile i was just in chile chile the big mac price is a very sensible 350 but the wage rate is only 2.15 cents an hour so they only have big macs per hour of 0.61 this is a sad case this is venezuela uh at the last time we had 2015 we collected the data the the because of the exchange rates uh control wage rates were 17 cents an hour big mac price was really cheap 67 cents big max per hour was .25 that clearly could not last in now all all mcdonald's are gone in venezuela uh let me show you a picture in 2007. i picked this example because 2007 is before the financial crisis this is when things were going going well and the way i've constructed this index is the first column has the i've there are too many countries you cannot display 65 countries so i've made continents special little continents uh and you'll see these are the more or less the largest countries uh and then i have the rest of asia i put eastern europe all together and i put western europe all together sorry about that and then the middle east where it's extremely difficult to collect data largely because many of the workers there are not indigenous they they come from abroad and it's hard to know exactly what that makes and then there's there's latin america uh which is probably too big as well you can see the general level of prices uh the us at that time was 733. uh only japan was was uh higher you can see uh typical china's at 81 cents uh and so on uh now in the second column i've taken the ratio of the country in question and do this many times so you have to pay attention i've taken the ratio of the country in question to the u.s just to make it easier to understand so the canadian muck wage was 93 percent of the us mcwage close russia was 32 south africa 23 china 11 india 6 japan almost exactly the same rest of asia 14 eastern europe 25 western europe 29 more than in the us and latin america 14 the big mac prices in the fourth column are in the third column over here that's in that's in dollars again uh and you can see that as expected the high wage countries have higher prices that's the loss of samuelson effect the big mac is partly tradable and partly non-tradable so you can't it's not exchange rates that explain these prices it's the fact that part of the product is made with non-tradable good and that non-tradable good is labor and in the right-hand column you have big macs per hour so you can see the us was at a high of 2.4 canada 2.2 russia a little over one western europe at 2.2 uh and you can read the rest of these off of here now what i want to show you next well there's a picture i couldn't resist this because this is actually from the internet so that that is an advertisement for mcdonald's in china and that's the advertisement for mcdonald's in the us it's if there's more stereotypical picture i do not know what it is and that's how many big macs you get in each place uh this is a picture that tries to give you an idea of the world distribution uh of uh the dollar milk wage rate so what i've done is i've i've the the wage rate runs along this axis and then up here is the is a is a population uh ratio right so it's the fraction of the world's population so you can see the fraction of the world's population that's at or below 250 an hour is very very high so generally speaking the the vast majority of the world which of course is in india and china has wage rates that are far below 2.50 cents an hour um this this is my attempt to show you some growth this is very good news i was so happy when i first did this from 2000 to 2007 so you can see what i'm doing i'm taking 2000 up to the before the financial crisis it's a little bit hard to understand this unless i explain it to you so the first the first the first column shows you uh the wage rate uh in the u.s in 2007 compared to 2000 so 1.13 means that the money wage increased by 13 in the us uh 50 in canada look at russia 460 china 92 india 57 japan a drop of 5 now what i've done is to put the take the ratio and then we'll take that number and use it as the denominator and then divide by each of these so i'm showing you what the growth is like in each of these countries compared to the u.s so needless to say the us is one uh the canadian case they grew by 34 34 more than the growth rate the growth rate there was 34 more than it was in the us the russian growth rate was four times as high as the u.s growth rate chinese one was 70 percent higher indian 40 percent higher and the japanese were dropping by 15 so i do the same thing with the big mac price ratio so that's the growth in the price of the big mac over there 21 in the us so you can do the comparison yourself you can see that because the price of a big mac went up more than the price than the wage rate the big mac per hour ratio declined so compared to its level in 2007 and 2000 the big max per hour went down the real wage in the u.s dropped by seven percent uh the real wage in canada dropped by eight percent the real wage in russia went up by 250 percent this is the source of mr putin's power this man has presided over the greatest growth in russian history from the point of view of any ordinary russian he's a genius it should never be forgotten the man has to be anybody who can do that for any country will be incredibly popular i could not even conceive of the us having a growth rate of 250 percent the real wage over less than say 100 years china 60 percent faster than the us india 53 percent and japan more or less stable so this is a great period in a way the poorest countries russia china and india grew rapidly fifty percent uh faster growth in in in china and india or more than that than in the u.s from 2000 to 2007 now the bad news uh this is the same analysis from 2007 to 2011. i put more countries in here because i had more countries you can see the us continues to decline canada has now declined there are only two countries on this chart uh other than eastern europe small growth there are only two countries that have significant growth russia and china the financial the recession basically the way we've adjusted to it everywhere is by reducing wages real wages have been in decline so the part about this that was the most depressing to me is india which people think of india as a fast-growing country but actually compared to population growth and in real wage measures uh that growth stopped several years ago and that it is true of course that because china is the largest country that poverty across the world continues to decline um and i showed you in fact the the wage rate has continued to increase up to 2016. uh let me show you a recent picture now it's a little busy the reason is because there are four countries but there are two measures of wage rates for each country you'll see that some of the measures start here that's because that's when we started doing it so the these are the countries for which we collect data regionally and it's big macs per hour uh it's so this allows you to compare the difference in the country as a whole with the big city a good example might be uh the us which uh is the gray line that's that's for the big cities the gray line and you can see the decline so it's going down been going down for a long time where's my red dot not as big a decline outside of the big cities uh you can see russia no matter how you measure it higher in this measure no matter how you do it you can see they have started to decline but they've had growth for a long time a little bump they're here but they're close for a long time china of course is much lower but it has continuous and continuous growth and then there's india which is basically tanked so this is really totally up to date so that's what's happening in the world um i think many people are surprised to see that the russian real wage could get as high as the us big shock to most americans but it has been a phenomenal growth period for the russians and despite the oil price collapse you don't see any collapse of real wages here it's a decline substantial decline but no collapse okay let me take a few minutes to talk about boss of samson so i mentioned this business about this is to explain more about why exchange rates can't be expected to adjust these things so what i've done is to show you on this axis uh the wage rate and on this access the big mac price and you see they go up but they don't go up linearly they go up this way that's the so-called bolas of samus in effect it's it implies that real wages are that that labor is about a a third of the cost of the big mac this is a very nice picture i love this picture this shows you the growth rate in big macs per hour compared to the level of the big mac per hour in 2009 what's what does it show well it shows that the high-wage countries have had the lowest growth rates the law these two there by the way the population size is the size of the ball you can see india and china have done better so the low wage the growth rates the us has had a substantial decline in real wages as have many other rich countries so this this is what we would hope to see if there's leveling in the world if the if the poor countries are catching up to the rich countries so it's the growth rate measured against the level initial period i wanted to show you this because it i wanted to show you that how it is that ppp is purchasing power parity i want to show you that so this is an incredibly cheap way to collect data on gdp a first cut measure of the gross domestic product can be obtained but is finding out how much somebody makes who works in the mcdonald's amazingly simple thing the places that are way off are the ones that have the very high minimum wages they can hear but generally speaking that's because they've re used the minimum wage to redistribute income for better or worse but the point is that generally speaking you can use that this is total factor productivity as measured from the pinworld table versus my measure of total factor productivity uh based on wage rates that mcdonald's adjusted for capital labor ratios and you can see they're very highly correlated do you want to know how well how productive a country is the simple way to find out is look at how much someone in a mcdonald's makes doesn't take much work uh i want to say one final thing uh and then and then we'll wrap this up this is a very provocative point i'm gonna make but far and away the cheapest way to make poverty go away is to move people uh if you went to michael kramer's talk which was very interesting you can see that making less developed countries become developed is very very difficult it takes a long time we never know exactly why it happens there's one way for sure that you can make very poor people richer move them to rich countries a mcguage is the job that anybody can do these are people who are doing the same work in every one of these countries if they can do it in china they can do it in the u.s they don't need very many words of english especially now that we're using touch screens so a tiny amount of english can get you there one of this is very provocative point because it's basically a pro argument for migration if you secure the world's poverty the simplest way to do it would be to move people from the poorest countries to the richest countries now i realize this is extremely provocative because there are many other downsides but the upsides are simply phenomenal in size uh this is work based on migration rates that have actually taken place uh let me show you this picture first the the bottom axis has the difference in wage rates between about 2 000 comparisons so i compare wage rates uh the the richer country to the poorer country uh across many you can imagine how many comparisons there are right if i have 70 countries i can compare each of the one country to all of the other 70. now i can take the remaining 69 comparing me to the rest so there are many many comparisons and these are the actual migration rates up on this axis you can see the interesting thing is there's no correlation between how much migration there is and what the wage gap is it isn't the case that migration rates are higher when the wage gap is higher this suggests that most migration is not not at least in any simple way totally economically determined but what it does do is this it says suppose i take the observed migration that we've seen and ask how much welfare how much money how much gdp was created by those movements now here i have a comparison of someone who does exactly the same thing in each country it's not like better migrants compared to lesser migrants these are just the same they're doing exactly the same thing so i solved the selection problem completely i'm giving this as kind of an example i solved the selection problem completely these are no better or worse no selection issues here at all i can compare people who moved and asked the question of how much money is involved well this this is in dollars right so you can see if you look i aggregate the number over here um you can see that the numbers are simply phenomenal uh so it's like uh four million dollars an hour that was created by the migration that took place in 2006. five four five million dollars per hour uh if i use more country repairs it can go up to 9 or 10 million dollars per hour that's added gdp to the world add the gdp to the world shows you the simplest way of all to make the world more equal is just to move many of the poorest people into the richest countries the alternative of course is to try to create richer countries where people already are that's slow and in some cases goes backwards uh in any event this is this is the provocative pro-migration argument that's really all i have to say but i do have a closing slide and you may if you like you may yell out where you think each of these mcdonald's is i think that's china we know that one good guess thank you very much unfortunately we haven't got time for further questions we have to leave the room therefore i would like to thank wholeheartedly uh professor for the fascinating and very unusual presentation thank you very much and i wish you a fruitful day thank you very much you