|
Hong Kong9151 Posts
Inflation and Exchange Rates
An examination into the failures of methodology commonly exhibited by the construction of competitive gaming earnings data aggregates.
Introduction
One of the opportunities that the longevity of the competitive gaming scene offers us as observers is the ability to look into the earnings of our best players through their performances in gaming competitions over time. A number of people find it worthwhile to compile statistics in this regard, and there are even websites dedicated to compiling these statistics.
The problem is that it appears none of these works actually do a proper job of accurately determining a player’s earnings over time. This is because they ignore the concepts of adjusting for inflation and accounting for historical exchange rates.
Definitions and Significance Inflation
Inflation is an economic concept describing a rise in the general level of prices of the goods and services in an economy over a period of time. As the price level rises, each unit of currency can buy fewer goods and services. In short, a set denomination of currency that you may have had in the past will not have the same purchasing power as that same denomination of currency you may have in the present, or, indeed, in the future. [1]
For example, a hamburger from McDonalds in the year 1955 was priced at $0.15 to buy. [2] In the year 2012, with the value of the American currency being what it is today, that same hamburger will end up costing $4.20 to buy. [3] [4] Over the last 57 years, the purchasing power of the individual dollar has decreased substantially.
If, on the other hand, the price of goods and services goes down over a period of time, the effect that this has on the purchasing power of an individual unit of currency is called deflation. As the prices of goods and services within the basket of a given economy declines, an individual dollar becomes worth more, as it can begin to purchase more and more goods relative to the decline in prices.[5]
Why is this important to our analysis? Simply put, someone who nominally wins $1,000 in the year 2003 (nominal meaning unadjusted for inflation) did not win, in real terms (having adjusted for inflation), the same amount of money as someone winning $1,000 in the year 2012. Failing to account for this difference in purchasing power is a fundamentally debilitating error in any statistical compilation of earnings over time.
Exchange rates, and historical exchange rates
An exchange rate, established between two currencies, is the rate at which one currency can be exchanged for another. For example, at the time of this writing, if I wanted to change 100 USD to its equivalent in South Korean Won, with the exchange rate being 1 USD = 1,112.42 KRW, I would end up with roughly 111,242.14 KRW. [6]
It must be understood that currency exchange operates in a market. The quoted exchange rate is an estimated middle ground between the level at which sellers of a certain currency are willing to trade an individual unit of currency (the Ask), and the level at which buyers of a certain currency are actually willing to pay for that same individual unit of currency (the Bid). And there have to be buyers and sellers of any currency in the first place for there to be liquidity in the market, which means that enough action is going around that the sale of an asset such as a unit of currency will not cause a significant movement in price and the corresponding change of value attendant with that. [7] [8]
For purposes of simplicity, we are going to assume that conversions between currency happen cleanly, with absolute guarantee of liquidity in the market, and that the sum total recieved at the end of an exchange is untouched by fees or taxation. In short, we are going to assume that the quoted exchange rate can be used as a simple conversion metric, even though it is not a simple conversion metric in reality.
The complicating factor in all of this is the fact that the exchange rates between currencies remain ever-changing. The exchange rate between the United States Dollar and the South Korean Won on 25 September will not be exactly the same 24 hours later on 26 September. Why is this a problem? It means that in converting historical earnings values, we cannot rely on current exchange rates to determine their equivalent value in another currency. Not just the exchange rate is wrong in this situation, but, once again, inflation between the two currencies at two given points of time are different as well.
Okay so...
What does this mean? It means that any accurate determination of earnings over time must account for inflation. Any currency conversions done must take into account the exchange rate on the day most relevant to the event at which money was earned, as well as for inflation at the end.
Proving that a problem exists
esportsearnings is a website that seeks to aggregate earnings statistics across a wide variety of competitive gaming titles, including StarCraft, Counter-Strike, Dota, League of Legends, and fighting games. To do this, they appear to scrape information from community websites such as Liquipedia, TLPD, and Shoryuken to gather earnings data. [9] Other than claiming to transform currencies into their equivalent US Dollar values at an exchange rate relevant to the date a prize was awarded, no further transformation of values is undertaken by the site.
In making the following analyses, I have relied on a number of resources, namely: historical foreign exchange rates provided by the United States Federal Reserve, [10] XE, [11] and the inflation calculations provided by the determination of the Consumer Price Index (CPI) by the United States Bureau of Labor Statistics. [12]
Lee Yoon Yeol, also known as NaDa, should need little to no introduction to people on this forum. His achievements in Brood War make him a serious contender for being one of the best of all time at that game. His achievements in StarCraft II brought him a new relevancy at the dawn of a new progaming era. That NaDa’s professional gaming career lasted so long offers us the ability to see the effects of inflation and historical exchange when looking back at his earnings over time.
For a more clear view of these data, click on the image for a larger version.
What these data point towards is twofold: (1) there is a difference in how the website calculates historical exchange rates, but that difference is somewhat negligible; what is not at all negligible is (2) the fact that there is a nearly $20,000 difference between the inflation adjusted rate and the non-inflation adjusted rate.
Failing to account for inflation causes severe underestimation of the value of earnings in competitive gaming in the last decade, privileging more recent results at the expense of the old.
Because of the ubiquitous necessity to deal with historic currency exchange in NaDa’s earning history, I moved onto another figure in competitive gaming who would have less currency exchange to deal with in his total.
Shane Hendrixson, also known as rapha, is a prolific professional Quake player who has won multiple major events throughout his long career. Because the majority of these events took place in the Western world, currency exchange factors very little into determining rapha’s earnings totals. [13]
For a more clear view of these data, click on the image for a larger version.
The difference between the stated earnings value [14] and the inflation adjusted earnings value seems to be far less here relative to NaDa’s example, but I contend that it is still significant. The relative difference in change between the two variables between rapha and Nada, I think, can be attributed to the fact that a majority of rapha’s victories are actually somewhat recent, relative to other progamers. To put to rest any lingering thoughts of, “maybe this is not the biggest deal,” let’s look at a final example.
Johnathan Wendel, also known as Fatal1ty, is listed as the highest earner in competitive gaming history on the website, and with good reason. The number of events this man has won over the past two decades is astounding, and he helped elevate the status of competitive gaming here in the West to new heights. [15]
For a more clear view of these date, click on the image for a larger version.
The difference between the stated earnings value [16] and the inflation-adjusted earnings is quite astounding. Nearly $100,000 of error is present in the stated earnings value. This is because of a point that I made earlier: failing to adjust for inflation makes any sort of medium to long-term aggregation of currency values quite worthless.
At this juncture, I would also like to call attention to the existence of another website purporting to aggregate earnings in competitive gaming, although their coverage is limited to StarCraft II competitive play. sc2earnings [17] is even less transparent than esportsearnings, providing no sources for their earnings values, and no hinting at any currency exchange methodology. For the purposes of the following analysis, I will assume that all values shown are correct, and that the values in dollars are legitimate conversions from the original currency to the United States Dollar at appropriate rates based on the date of the awarding of prizes.
In examining sc2earnings, I took Jang Min Chul, also known as MC, to be the most relevant subject for analysis. His StarCraft II career has spanned since the beginning of play to the present day, offering some inflation adjustment analysis to be done to his earnings figures over time.
It is fairly certain that sc2earnings also does not adjust their values for inflation, as there is a marked difference between the stated earnings total [18] and the inflation-adjusted total, which is significant.
Implications and Final Thoughts
Unless an esports earnings resource adjusts for inflation and accounts for historical exchange rates, the data they provide is worthless for any serious mid- to long-term analysis. Failing to account for inflation seriously underestimates the worth of events and earnings made in years past, and privileges more recent results instead. Continuing to rely on faulty resources and methodologies leads to the misleading of people doing research into the viability of competitive gaming. If these faulty resources are being used to make economic decisions, I think that decision makers should think twice and critically evaluate their sources of information moving forward.
Go ahead and comment below and share your thoughts on the matter.
Footnotes and References:
- "Inflation". Wikipedia. 19 September 2012. Retrieved 28 September 2012.
- Lynne Olver (23 September 2012). "Food Timeline FAQs: Historic Food Prices". Food Timeline. Retrieved 28 September 2012.
- "Big Mac index". The Economist. 14 January 2012. Retrieved 28 September 2012.
- This, of course, assumes that the physical makeup of the hamburger itself has not changed significantly in such a way so as to affect a change in the intrinsic value of the hamburger meal itself. Thus, it is assumed that the hamburger you can buy in the year 1955 is the same as the hamburger you can buy in the year 2012.
- "Deflation". Wikipedia. 27 September 2012. Retrieved 28 September 2012.
- "Exchange rate". Wikipedia. 19 September 2012. Retrieved 28 September 2012.
- "Market liquidity". Wikipedia. 22 September 2012. Retrieved 28 September 2012.
- "Bid–offer spread". Wikipedia. 3 September 2012. Retrieved 28 September 2012.
- "e-Sports Earnings". e-Sports Earnings. Retrieved 28 September 2012.
- Board of Governors of the Federal Reserve System (24 September 2012). "Foreign Exchange Rates - H.10". Retrieved 28 September 2012.
- "Current and Historical Rate Tables". Retrieved 28 September 2012.
- United States Department of Labor Bureau of Labor Statistics. "CPI Inflation Calculator". Retrieved 28 September 2012.
- "Shane Hendrixson". Wikipedia. 19 September 2012. Retrieved 28 September 2012.
- "Shane 'rapha' Hendrixson". e-Sports Earnings. Retrieved 28 September 2012.
- "Johnathan Wendel". Wikipedia. 13 August 2012. Retrieved 28 September 2012.
- "Johnathan 'Fatal1ty' Wendel". e-Sports Earnings. Retrieved 28 September 2012.
- "SC2 Earnings". Retrieved 28 September 2012.
- "SC2 Earnings - Player Details". Retrieved 28 September 2012.
Previous Blog:On Tournament Design
   
|
|
Hong Kong9151 Posts
thanks, I'll try to clarify those terms a bit when they first appear
|
congratulations on passing your economic 101 and 102 exams.
|
Good write up, thanks for the info. There isn't really too much to add, except - I understand why you have posted it here on TL, have you raised this with the sites you mentioned?
|
Magic Woods9326 Posts
I was promised a picture of NaDa what ever happened to that?
|
Interesting, there is also the issue of what inflation index do you use. Is the inflation right now at 1.6%? The official US numbers say so, but when I go to the grocery store, I see a different picture..
|
Interesting topic to write about, I just recently began AP Economcs in my school and really enjoy the class as well as this article!
|
Hong Kong9151 Posts
On September 29 2012 19:50 Epoxide wrote: I was promised a picture of NaDa what ever happened to that?
Opps, sorry!
On September 29 2012 20:07 brolaf wrote: Interesting, there is also the issue of what inflation index do you use. Is the inflation right now at 1.6%? The official US numbers say so, but when I go to the grocery store, I see a different picture..
I use United States CPI (Consumer Price Index) as calculated by the US Bureau of Labor Statistics because my analysis is in United States Dollars. CPI shift is currently at +0.6% since last month according to that metric. Overall change for the year is around +1.7 percent, this being the overall inflation metric for the economy. Source (PDF)
Keep in mind CPI is a broad basket, and local prices may be way more inflated than on the overall national level of analysis.
|
Magic Woods9326 Posts
On September 30 2012 01:26 itsjustatank wrote:Show nested quote +On September 29 2012 19:50 Epoxide wrote: I was promised a picture of NaDa what ever happened to that? + Show Spoiler +Opps, sorry! On September 29 2012 20:07 brolaf wrote: Interesting, there is also the issue of what inflation index do you use. Is the inflation right now at 1.6%? The official US numbers say so, but when I go to the grocery store, I see a different picture.. I use United States CPI (Consumer Price Index) as calculated by the US Bureau of Labor Statistics because my analysis is in United States Dollars. Inflation is currently at 0.6% according to that metric. Lets say I read it after you posted the picture
|
United States7483 Posts
Just to point out about his explanation of inflation: you can't simply pick and choose Macdonald's hamburgers as your basis for measurement of inflation out of the blue, you need to account for variable bias in the selected statistic. Maybe in the starting year there was an enormous surplus of ingredients which is no longer available, maybe there is now a shortage that isn't normal, etc. Maybe Macdonald's is deliberately overcharging for their burgers compared to what equilibrium price should be.
Typically inflation is measured across multiple different goods throughout the time period, or goods more or less free of variable bias are used.
|
Hong Kong9151 Posts
On September 30 2012 02:00 Whitewing wrote: Just to point out about his explanation of inflation: you can't simply pick and choose Macdonald's hamburgers as your basis for measurement of inflation out of the blue, you need to account for variable bias in the selected statistic. Maybe in the starting year there was an enormous surplus of ingredients which is no longer available, maybe there is now a shortage that isn't normal, etc. Maybe Macdonald's is deliberately overcharging for their burgers compared to what equilibrium price should be.
That isn't my measure of inflation, it's an example of inflation. And sure, they do have some price-setting power perhaps, but if anything they undercharge for their burgers as part of a deliberate anti-competitive practice.
Inflation is calculated by measuring CPI, and I use that metric throughout.
|
very true. I think they should probably display people's earnings through adjusted inflation if we want to continue with e-sports into the long term like +5 +10 years
|
So just to clarify, you're stating what they would have had to earn today (as the adjusted for inflation) in order for it to be equal with what they earned in the past? That is, for it to have the same purchasing power?
Because at first glance it seems like you're saying the money that they earned in the past, valued today, is worth more after inflation - but I would think that "adjusted for inflation" would decrease the value of what they had won rather than increase it.. So as an arbitrary example, if they earned $1000 dollars at a tournament in the year 1995, then today that same $1000 would have the same purchasing power as $500. And I guess this is all relative to a specific year, otherwise you can't make sense of what the baseline value of money is supposed to be and what increase or decrease in value means.
|
Hong Kong9151 Posts
On October 01 2012 01:27 radscorpion9 wrote: So just to clarify, you're stating what they would have had to earn today (as the adjusted for inflation) in order for it to be equal with what they earned in the past? That is, for it to have the same purchasing power?
Because at first glance it seems like you're saying the money that they earned in the past, valued today, is worth more after inflation - but I would think that "adjusted for inflation" would decrease the value of what they had won rather than increase it.. So as an arbitrary example, if they earned $1000 dollars at a tournament in the year 1995, then today that same $1000 would have the same purchasing power as $500. And I guess this is all relative to a specific year, otherwise you can't make sense of what the baseline value of money is supposed to be and what increase or decrease in value means.
These two paragraphs are not mutually exclusive, so yes. I don't always assume the past value is going to be greater in today's dollars, because deflation (or negative inflation) exists as well in certain economies. The larger point is that unless you adjust for inflation (or deflation), displaying earnings statistics over the medium- to long-term and presenting them as fact (and giving a non-inflation adjusted sum of the values) commits a fundamental error.
|
Israel2209 Posts
radscorpion9, I think one of the main points is that these lists published by websites such as esportsearnings and sc2earnings are simply invalid. You cannot add the $1 you earned in 1990 and the $1 you earned in 2012 and say that you earned "$2" from 1990 to 2012, that's the first things these sites do wrong. Second, they are comparing "$2" that you earned in 1990-2012 with the "$3" that I earned between those years. They are comparing two figures which are wrong to each other. For anyone that knows anything about economics/finance, these mistakes are so rudimentary - it is just silly. Imagine asking someone to weigh the total amount of apples he has, and he weighs one apple then multiplies the amount of apples he has with that weight. Did the number he reach have any meaning? Would you post it on your site?
|
On October 01 2012 01:27 radscorpion9 wrote: So just to clarify, you're stating what they would have had to earn today (as the adjusted for inflation) in order for it to be equal with what they earned in the past? That is, for it to have the same purchasing power?
Because at first glance it seems like you're saying the money that they earned in the past, valued today, is worth more after inflation - but I would think that "adjusted for inflation" would decrease the value of what they had won rather than increase it.. So as an arbitrary example, if they earned $1000 dollars at a tournament in the year 1995, then today that same $1000 would have the same purchasing power as $500. And I guess this is all relative to a specific year, otherwise you can't make sense of what the baseline value of money is supposed to be and what increase or decrease in value means.
Your line of thought is folly.
It's reasonable to assume that they spent said money, and didn't simple save it with zero interest. For example the house that Mvp bought was used with money that was worth more 2 years ago than what it is worth today, but the house remains at the relative same value.
Therefor, his overall value from 2 years ago has increased, not declined assuming the price of his house follows that of inflation (which we have to assume, and which is most likely).
|
Hong Kong9151 Posts
Whether they spent the money or invested it with some return doesn't really matter if the point is to show earnings at X tournament in 2005 and then add it to earnings at Y tournament in 2012 in order to display a player's n earnings total.
The values can't be added together because they aren't in the same dollars when accounting for inflation. And because the frame of analysis is from the year 2012, showing X tournament's dollars in that year's dollars and not the current dollars is misleading.
|
On October 01 2012 05:26 Figgy wrote:Show nested quote +On October 01 2012 01:27 radscorpion9 wrote: So just to clarify, you're stating what they would have had to earn today (as the adjusted for inflation) in order for it to be equal with what they earned in the past? That is, for it to have the same purchasing power?
Because at first glance it seems like you're saying the money that they earned in the past, valued today, is worth more after inflation - but I would think that "adjusted for inflation" would decrease the value of what they had won rather than increase it.. So as an arbitrary example, if they earned $1000 dollars at a tournament in the year 1995, then today that same $1000 would have the same purchasing power as $500. And I guess this is all relative to a specific year, otherwise you can't make sense of what the baseline value of money is supposed to be and what increase or decrease in value means. Your line of thought is folly. It's reasonable to assume that they spent said money, and didn't simple save it with zero interest. For example the house that Mvp bought was used with money that was worth more 2 years ago than what it is worth today, but the house remains at the relative same value. Therefor, his overall value from 2 years ago has increased, not declined assuming the price of his house follows that of inflation (which we have to assume, and which is most likely).
Oh no, I would never propose something so silly . I am merely using it as a hypothetical example to help explain the effects of inflation strictly in terms of the valuation of money, I am not saying that's actually what pro-gamers do. In the table he is directly comparing earnings from several years ago to today, so I was doing the same thing but using my own hypothetical case.
But anyway thank you for the responses. I agree with the point of the OP, that you can't simply compare earnings from one year to another without adjusting for inflation, I was just curious as to the finer point of which way inflation is affecting their earnings - and it appeared to me that it should be the opposite way. It seems like my first description was on target so no further questions!
|
While all of these are relevant points, the entire idea of earnings comparisons are misleading.
A lot of these guys earn money in and live in entirely separate places from each other so purchasing power parity plays a big part yet can't really be taken into account in these tables.
Just some food for thought.
|
|
|
|