Guild Wars 2

Introduction:
A long time ago, someone made a very interesting post here on reddit, about the inflation in Guild Wars 2. The research is no longer available and both the website and youtube video are gone. For a while, I’ve wanted to update the research and it’s finally done.
I expect the launch of HoT to have a big impact on the economy of the game and I want to have some actual data in order to see what will happen, from an economical point of view. The plan is to regularly update the spreadsheet, so that all of you will be able to follow the changes in the inflation.

https://docs.google.com/spreadsheets/d/1fVpco8hEdap2AtcB3NCbMovHtrPu-mbrYSLqwbue4fg/edit#gid=1079668078

NB: Inflation can be a very complicated topic, and I’ve deliberately left out a lot of things in my explanations in order to keep it as simple as possible.

What is inflation:
Definition: Inflation is a sustained increase in the general price level of goods and services in an economy over a period of time.
As the definition explains, inflation is the increase of the general price level (deflation if it’s a decrease). This means that the 10 gold you had last year, will now be worth less. You will not be able to buy the same, since the value of the gold is decreasing over time.
Inflation is there from a few different reason. It can be the cause of shift in the demand and supply. E.g. the introduction of ascended armor caused the prices of many items like silk and mithril to rise. This is due to the higher demand of these items.
Another reason for inflation is the increase of gold in the economy. In the beginning of the game we were farming CoF P1 (a lot) and got a few gold per hours. Today we can farm Silerwaste and easily get 10-20 gold and hour or just do a dungeon tour. Since the players have more gold, the prices will increase. The games does not have that many goldsinks and we have an increase in the gold supply. As more player will try to buy the same item, the prices will rise.

Method of calculation inflation:
In order to calculate the inflation in the game I’ve used a Consumer Price Index (CPI). I picked 107 items from the game and created something called a “basket”. Each item is then given a weight depending on how important, that specific item is for the economy. This basket is constant over time and the items and their weight is always the same.
For all these items I’ve recorded the price at the beginning of every month (sell prices in silver). This way we will be able to see how the cost of exactly the same items differ over time. I’ve created an index starting from October 2012 and calculated the monthly inflation for the past three years based on this.
I chose not to use the data from September due to the very unstable prices and because some of the items were not even being sold on the trading post at the time.

How to interpret the numbers:
The inflation research is based on index numbers. For index numbers you set a specific time and the numbers will always refer back to this point. An index of 194, means that the prices are 94-percent higher compared to the first month in the index.
Other than the index numbers I’ve calculated the monthly change in the inflation rate. This gives an indication of how the inflation moves up and down (deflate) over time. Both of these numbers are represented on timelines for a more visual representation.
I decided to calculate the inflation for a some group of items as well, like precursors. For some of these indexes I’ve done the calculations with multiple base months. This is again due to the very unstable prices during the first few months of the game.

My predictions for the launch of Hearth of Thorns:
To say anything about what the economic effects of Hearth of Thorn will be, is very kittene thing that I can say for sure, is the fact that we have a huge influx of P2F players and I will assume that a lot of old players will be returning to the game ones that expansion is out, if they are not already back.

EDIT: The declining deflation is caused by the introduction of maps like silverwaste and Dry top, where players farm a lot. Insteadt of farming raw gold, we actually gain most of the rewards in terms of materials. This causes a shift in supply and causes deflation.

From the declining deflation for the last six months, my guess is that that a lot of players have been saving their gold for things like the new legendries and the other new items coming with the expansion.
From the previews of the new raids, we now that berserker will not be the only viable armor stats and the meta will change to include condition damage (maybe for FotM as well). Many will have to craft new armors and the demand for these items will go up. All these things will push towards a higher inflation, due to a higher demand on the items in the basket.

On the other hand we know that the supply side will change as well. The precursor crafting is coming and from the new map reward system we will get alternative ways to farm a lot of very valuable items. We will get new PvP rewards and the new rewards from FotM 100 and raids will have an influence as well (the non-account bound). These items will push towards a lower inflation, because the supply curves will shift.
At the moment we are already seeing a rise in the inflation and we are not far from the last 2014 number . From all the things we know will be introduced with HoT, I believe that this trend will continue and the inflation will increase even more. This is just my specilations, we wil have to wait and see whatever I’m right or worng.

Conclusion:
So if we calculate the yearly inflation we get these three numbers. For the first year we had a 71,406 % inflation, the second year we reached 81,8668% and for the last year we have had a 6,9829 % deflation (-7%). To give you all some perspective of how high the inflation in the game have been, we can use an example from the real world. For the past 5 decades in the USA, the average annual inflation have been between 1,86 % and 5,82 %. Reaching more than 80 % is very unusual in the real world. The real world is also much more complex and the governments and national/central banks are doing what they can in order to keep the inflation as low as possible (while maintain a low unemployment rate). Areanet simply don’t have the same tools at the disposal in order to stabilize the economy and I would assume that most game economies would look this (Edit: read my 3rd edit where I elaborated a little on this comment).

I hope some of you actually made it this far, I know it was a long read. This is something I find very interesting, so feel free to ask questions or comments below. In the spreadsheet you will be able to find all the numbers and graphs. Thanks for reading!

TLDR: The prices of things go up. You would now need 190-percent more gold in order to buy exactly the same things that you could have brought back when the game launched.

EDIT1: Changed a few words.

EDIT2: As someone pointed out, I made a mistake. The reason we have had a deflation the last year can be because of the introduction of Silerwaste. A lot of players are farming there and insteadt of creating raw gold like dungeons do, we get materials and items back. This actually causes a shift in the supply curve which again causes a deflation. It’s an easy way to get materials and the fact that we have more items comming into the economy, the prices are going down.

EDIT3: As a few people have point this out, i said that Arenanet don’t have the same tools at their disposal at the central/national banks. What I mean by this is simply that they have different tools at theirs disposal, not that Arenanet have no control. The goverments can change things like exchange rate, govermental spendings and can more easily moniter the money supply.

EDIT4: https://www.guildwars2.com/en/news/the-guild-wars-2-heart-of-thorns-economy/ A new blog post just came out, regarding some changes for HoT. From what I can see, Arenanet are actually going away from rewarding gold and to materials as well. From en economical point of view, this is a very good change, and could be a way to lower the inflation significantly in the future.
Original reddit post: https://www.reddit.com/r/Guildwars2/comments/3obkcl/guild_wars_2_inflation_research/
Decided to post it here as well in order for more people to see it!

This is a very misleading post… based on your actual data the inflation rate is almost 0, not 80%.

You just included the first few months where you can’t possibly calculate inflation to make it look bad… After the first year the economy was fully stabilized and there has been almost no inflation since then.

ZudetGambeous.9573:

This is a very misleading post… based on your actual data the inflation rate is almost 0, not 80%.

You just included the first few months where you can’t possibly calculate inflation to make it look bad… After the first year the economy was fully stabilized and there has been almost no inflation since then.

That’s far from true. I agree that the economy was not very stable at launch, which is why I left out that first 1½ in the research and did create multiple indexes for a lot of the calculations. But this changes nothing in terms of calcualting the yearly inflation.

01/10/2013 – 01/10/2014: (01/10/2014 – 01/10/2013) / 01/10/2013 = (7.249.787,59 – 3.986.317,36) / 3.986.317,36 = 81,87%

Even if I did do it as you said, I remove the fist year of the game, the inflation for the 2nd year is still 81 %,

This is not a post to blame the economy or say how bad it is, this is simply just showing some data. I agree that the inflation could be lowered with some actual gold sinks, but that’s not what I’m trying to show.

DennisChrDk.9823:

ZudetGambeous.9573:

This is a very misleading post… based on your actual data the inflation rate is almost 0, not 80%.

You just included the first few months where you can’t possibly calculate inflation to make it look bad… After the first year the economy was fully stabilized and there has been almost no inflation since then.

That’s far from true. I agree that the economy was not very stable at launch, which is why I left out that first 1½ in the research and did create multiple indexes for a lot of the calculations. But this changes nothing in terms of calcualting the yearly inflation.

01/10/2013 – 01/10/2014: (01/10/2014 – 01/10/2013) / 01/10/2013 = (7.249.787,59 – 3.986.317,36) / 3.986.317,36 = 81,87%

Even if I did do it as you said, I remove the fist year of the game, the inflation for the 2nd year is still 81 %,

This is not a post to blame the economy or say how bad it is, this is simply just showing some data. I agree that the inflation could be lowered with some actual gold sinks, but that’s not what I’m trying to show.

Interesting research, but you have 1 critical flaw. Your basket of items used is flawed. Price inflation in the real world is based on common household items or other commodities depending on which inflation calculation you look at (overall, weekly, etc.). Personally I’d like to see a calculation based on necessity items that players require weekly. That would give a way better representation of what type of inflation is felt by players.

The first mistake is including precursors. Those are not wide spread common items (aka no normal household items). Those are luxury high value items (aka luxury cars, planes, yachts, etc.).

Your second mistake is not to factor for game mechanic changes. Cooking ingredients went through some changes of availability. Your inflation research makes it look that they have grown in price in a similar fashion as other items, where it is mostly based on change of aquisition. Again problematic when considering that these items again are not mandatory and most of them can get skiped while still getting similar effects with exchange goods (leveling cooking, similar buff food).

Finally, I don’t agree to some of your weighting choices. especially not with your weighting of ectoplasms with 10,000 units. Ectos are the very basis for a huge part of the general economy. You have weighted them way to low in my opinion.

Still cudos to the effort, interesting numbers.

EDIT:
Looking over the numbers again, the main inflation driving force is the increase in gold to gem prices. Not sure gem prices should be part of the ingame inflation calculation basket. Just about all of the gem items are luxury items, see my argument about precursors. But this can be debatable.

Wanna hedge against inflation? Buy gold ^_^ Seriously, we’ve been long overdue for ascended jewel crafting for some time.

The Tyrian gold marketplace is a very noisy place to look for the effects of currency quantity theory. Using arbitrary weights only increases the probability of generating noisy measures of inflation. Weights need to be based on real usage data. Have you considered using changes in buy and sell orders to make your weights more realistic. Of course, changes in buy and sell orders can’t be considered completely accurate representations of production.

On luxury goods:

Luxury goods aren’t used as part of real world indexes for many reasons, none of them relevant to Tyria. In the real world, luxury goods account for a tiny percentage of currency activity. In Tyria, that percentage is likely much higher. I wouldn’t be surprised if the production of luxury goods accounts for the majority of currency activity.

There are no goods that a Tyrian has to purchase every week. A Tyrian’s spending habits are dominated by gold price point compared to gathering price point and income; meaning every good is a noisy place to look for currency quantity theory. It is very inexpensive to remain relevant as a Tyrian hero, the cost of remaining relevant (equipped in exotic gear) approaches zero. An index that approaches zero is a terrible place to look for inflation.

Luxury goods are very relevant. Every character that I have lvled to 80 and equipped in exotics survived the experience and are now spending their time earning income to produce luxury goods.

We have to be willing to look for an objective measure of game wide inflation as well as consider which index is the best subjective measure of fair and equitable game design. Imho, that would be the index of crafting materials.

Inflation is a measure of the cost of affording the necessities of life: food, shelter, clothing, transportation.

• “Food” isn’t required in this game, so inflation = 0
• Shelter isn’t available; inflation = 0.
• Clothing: unlike RL, this is bought once and lasts forever (until you upgrade). No easy way to measure inflation for it.
• Transportation: cost of using waypoints has never changed. Inflation = 0.

tl;dr “inflation” isn’t a useful tool in measuring the difference in costs for players in 2012, 2013, 2014, and 2015.
————————————————

You could, instead, look at the total cost of acquiring certain useful and not-so-useful things. For example:

• Outfitting a new L80 in full, crafted exotics.
• Unlocking dyes for a 5-character account.
• Outfitting at least one character with an AR=70 set of gear.
• Completing any BL weapon collection.
• Completing Treasure or Exotic hunter collections.

Turns out: some of those have gone down in price and some have gone up.

Illconceived Was Na.9781:

Inflation is a measure of the cost of affording the necessities of life: food, shelter, clothing, transportation.

• “Food” isn’t required in this game, so inflation = 0
• Shelter isn’t available; inflation = 0.
• Clothing: unlike RL, this is bought once and lasts forever (until you upgrade). No easy way to measure inflation for it.
• Transportation: cost of using waypoints has never changed. Inflation = 0.

tl;dr “inflation” isn’t a useful tool in measuring the difference in costs for players in 2012, 2013, 2014, and 2015.
————————————————

You could, instead, look at the total cost of acquiring certain useful and not-so-useful things. For example:

• Outfitting a new L80 in full, crafted exotics.
• Unlocking dyes for a 5-character account.
• Outfitting at least one character with an AR=70 set of gear.
• Completing any BL weapon collection.
• Completing Treasure or Exotic hunter collections.

Turns out: some of those have gone down in price and some have gone up.

That is a very subjective definition of inflation, one that is completely removed from currency quantity theory. This from the US BLS website
http://www.bls.gov/dolfaq/bls_ques6.htm

I do agree with your second paragraph. The player base is varied enough to make individual indexes more relevant than an all goods index.

edit: Your definition of inflation is actually a description of an index.

If you all are so much kitten smarter than OP, maybe you could take the data that he spent a long time collecting and make something useful out of it.

I for one found in interesting.

Neural.1824:

If you all are so much kitten smarter than OP, maybe you could take the data that he spent a long time collecting and make something useful out of it.

I for one found in interesting.

It looks like most responders did exactly that, and as with all data sets, were able to come to different conclusions. Contradictory interpretations of the same information is normal.

Personally i would say the biggest “inflation” was after 11 months or so when we
got ascended weapon crafting and champion bags. Before i could farm around
maybe 2 gold per hour, after that it was suddenly 8-12 gold per hour mainly due
to increased prices of metal and wood.

In the end a lot of other stuff followed where the prices went for 4-5x higher
at that time, also the gold->gem exchange followed that way.

Funny is that ecto prices were already at 37s 6 months after release, when i
made my Vision of the Mists .. and they are still not much higher in price.

Illconceived Was Na.9781:

Inflation is a measure of the cost of affording the necessities of life: food, shelter, clothing, transportation.

I completely disagree with this. Necessities can inflate of course but they are not the sole measure of inflation. Inflation measures the cost of anything. Now I find it very disturbing all the greedy and heartless position had on inflation. Inflation it a tool used to interpret greater issues. This is a game, remember, that people play to escape the sucky real world. Luxuries need to be obtainable by all otherwise it frustrates people and they feel cheated by the time they invest in GW2. Exclusivity only creates a toxic environment. If this game is ever going to get back to its former glory it needs to think about all player experiences not just the hardcore or PVP player types. The inflation in this game proves why so many find this game unrewarding and discouraging. If I can’t obtain what I want with reasonable effort they I will just play another game that will treat me better. Inflation is generally used to show bad trends and is a warning that should not be ignored. If it is too high or even if it is too low it always translates to indicate something negative.

Greener.6204:

Neural.1824:

If you all are so much kitten smarter than OP, maybe you could take the data that he spent a long time collecting and make something useful out of it.

I for one found in interesting.

It looks like most responders did exactly that, and as with all data sets, were able to come to different conclusions. Contradictory interpretations of the same information is normal.

No one, including the OP, has access to the most important data in determining market-wide inflation: accurate production and usage data on which to base weights. The only thing we can objectively do is talk about methodology.

Sorry that I’ve not been able to respond untill now, have focused a little more on the reddit threadt (and I’ve not been home for most of the day), since it’s a little easier to respond to a single comment there.

You do all have some very valid points, and I guess you could discuss for hours about how excatly you should define the basket. It’s already something I’ve spend a lot of time on, but then again that was the hardest part of it. You haev to find some common ground between all different game types, and figure out what all players would buy.

In the end I decided to simply include some of the most demanded items used for cooking, crafting armor, weapons and legendaries and even gems. I found that these items represented what was being sold and brought in the economy the most. I would argue that the economy is centered around creating legendaries and ascended items, and it’s a goal that most players are working towards. If some players stick to excotic, the basket does include that with the tier 6 crafting materials.

In case you have not already checked out the reddit post, go look at it. Isaiah Cartwright have made a lot of comments about how they actualyl take all of this into account when they are balancing the economy, it’s defenetly worth a read!

Another thing. I’ve actualyl updated the spreadsheet alot today, and added a lot of new tabs and graphs. I’ll look into some of your sudgestions and do some more calculations

Beldin.5498:

Funny is that ecto prices were already at 37s 6 months after release, when i
made my Vision of the Mists .. and they are still not much higher in price.

Isaiah did actually explain this in the reddit post. They do center the economy around Ectos!

https://www.reddit.com/r/Guildwars2/comments/3obkcl/guild_wars_2_inflation_research/cvvvqdx

Gold to Gems shouldn’t be considered in the calculations at all. It’s entirely dictated by the imbalance in gems bought vs gems sold. Gems bought is dictated by gold availability and desirable items on the Gem Shop.

DennisChrDk.9823:

Sorry that I’ve not been able to respond untill now, have focused a little more on the reddit threadt (and I’ve not been home for most of the day), since it’s a little easier to respond to a single comment there.

You do all have some very valid points, and I guess you could discuss for hours about how excatly you should define the basket. It’s already something I’ve spend a lot of time on, but then again that was the hardest part of it. You haev to find some common ground between all different game types, and figure out what all players would buy.

In the end I decided to simply include some of the most demanded items used for cooking, crafting armor, weapons and legendaries and even gems. I found that these items represented what was being sold and brought in the economy the most. I would argue that the economy is centered around creating legendaries and ascended items, and it’s a goal that most players are working towards. If some players stick to excotic, the basket does include that with the tier 6 crafting materials.

In case you have not already checked out the reddit post, go look at it. Isaiah Cartwright have made a lot of comments about how they actualyl take all of this into account when they are balancing the economy, it’s defenetly worth a read!

Another thing. I’ve actualyl updated the spreadsheet alot today, and added a lot of new tabs and graphs. I’ll look into some of your sudgestions and do some more calculations

I did recall from the past that this had come up about ectos, hence my focus on them.

The basket is basically the means to an end, depending on which items are used to determine the inflation calculation define what result one gets. These are different approaches I see currently:

- Luxury items and expensive skins have grown in price, I’d likely agree (though so has the gold income and gold per hour). Thus inflation calculations based around these items, as well as gems, will reflect this I assume.

- my train of thought always has been though:“How much gold does one realistically need to hit the minimum gear required (exotic as is, with ascended weapons and trinkets) to participate in all PvE activities without being at a disadvantage.” This might rise depending on the necessity for ascendedd with HoT.

I’ll repeat myself here, great research and cudos for the work put into this.

DennisChrDk.9823:

ZudetGambeous.9573:

This is a very misleading post… based on your actual data the inflation rate is almost 0, not 80%.

You just included the first few months where you can’t possibly calculate inflation to make it look bad… After the first year the economy was fully stabilized and there has been almost no inflation since then.

That’s far from true. I agree that the economy was not very stable at launch, which is why I left out that first 1½ in the research and did create multiple indexes for a lot of the calculations. But this changes nothing in terms of calcualting the yearly inflation.

01/10/2013 – 01/10/2014: (01/10/2014 – 01/10/2013) / 01/10/2013 = (7.249.787,59 – 3.986.317,36) / 3.986.317,36 = 81,87%

Even if I did do it as you said, I remove the fist year of the game, the inflation for the 2nd year is still 81 %,

This is not a post to blame the economy or say how bad it is, this is simply just showing some data. I agree that the inflation could be lowered with some actual gold sinks, but that’s not what I’m trying to show.

I guess the biggest problem is the wild variation in the calculation.

If I look at 01/01/13-Today then the inflation over the last 2 years is only ~25%. I can pick any two arbitrary months to get anything from -20% to 50% in that same time period.

The best indicator of spending power is pretty much 100% tied to ectos. They are used in all armor, all weapons, and most other recipes. As long as you can buy ectos you can make everything you need. The ecto buying power has also remained remarkably stable over the last 3 years.

The problem is that many of the other goods you are tracking are based on changes to the game supply/demand curve, not inflation. Precursors are luxury items and have no place in an inflation calculation. Gold to gems are heavily dependent on gem store items and sales, not ingame inflation. If you remove these two erroneous items and weight ectos properly you will see a much different result.

ZudetGambeous.9573:

I guess the biggest problem is the wild variation in the calculation.

If I look at 01/01/13-Today then the inflation over the last 2 years is only ~25%. I can pick any two arbitrary months to get anything from -20% to 50% in that same time period.

The best indicator of spending power is pretty much 100% tied to ectos. They are used in all armor, all weapons, and most other recipes. As long as you can buy ectos you can make everything you need. The ecto buying power has also remained remarkably stable over the last 3 years.

The problem is that many of the other goods you are tracking are based on changes to the game supply/demand curve, not inflation. Precursors are luxury items and have no place in an inflation calculation. Gold to gems are heavily dependent on gem store items and sales, not ingame inflation. If you remove these two erroneous items and weight ectos properly you will see a much different result.

I’d wager the biggest problem is that the ig economy is controlled for the most part. That makes it vastly different from real world economies and real world inflation. Thus we cannot define ig inflation by the same parameters as rl inflation.

Take ectos for example……we have had major alterations to them.
Ecto salvage rates being bugged
Ectos made salvagable into dust
Introduction of world boss rewards
Game population magic find increasing
Guild event rewards…etc..

How is inflation measured by items when major alterations are made to said items’ demand and supply? This goes for most every item in the game, making rl economics moot in a lot of areas.

Behellagh.1468:

Gold to Gems shouldn’t be considered in the calculations at all. It’s entirely dictated by the imbalance in gems bought vs gems sold. Gems bought is dictated by gold availability and desirable items on the Gem Shop.

It was just more that some people think the prices there have exploded while i
for myself thought before ascended weapons i payed 30 gold for 1000 gems
that took me 15 hours farming at that time .. later when 1000 gems were at
150g that in the end only costed me 10 hours farming.

So alot in the end is relative from price to money that you can farm. In that
case there was for example also a big deflation for T3 armor since you can
get the money for it now much faster.

I don’t get what you people problem is.

There are 2 things that can happen.

A) To afford your stuff you are forced to buy gems and support A-Net
B) Less people have cool stuff and i can sell my stuff more expensive

As you may guess i am extremely fine with both

sorry, this is too much info.

so what is the average inflation per month? 10%?

Two things:

1) You really should not have gems included in your basket at all. It’s the equivalent of including Euros in a measure of US inflation – it’s not capturing what you want to capture at all and just throws off your measure.

2) Are you using any sort of intra-month price smoothing or are you just taking point estimates from the first day of every month? Unsmoothed values are going to show a lot of volatility that isn’t part of the real trend.

Otherwise you’ve done a pretty impressive job; people can quibble about what you’ve put into the basket and why, but this is more or less the process you want to follow to measure inflation.

Excellent post, nice effort for the data collection.
I think i remember this from the old bltc forums when it was posted first.

Back then and also now, I also dont think that the chosen baskets are ideal.
Most of the issues have already been mentioned, it mostly focusses on t6 items and luxury goods.

Crafting materials should be included of course but i think it would make more sense to index all kinds of lootbags, from kitten, instead of the crafting materials itself.

Every lootbag basically already represents the average value of the items inside, depending on their droprate. Of course this isnt an accurate summary because its being done via supply and demand on the tp but would still give better info than indexing only t6 common, fine and rare mats.

Nice effort. You have a great platform to set up indecies to model inflation.

I’d set up my indexes as

(1) High end WvW consumables. Basically foods and oils that serious WvWer pay for.
(2). Legendary weapon related markets: Rares, T6 mats, T5 materials, mystic coins…
(3) Non precursor, non collection exotics. Exotics that can’t be mystic forged and aren’t part of a collection.
(4) Set up a index immediately for any Pvp only drops
(5) Drops related to HoT precursor crafting and new Legendaries.

SkyShroud.2865:

sorry, this is too much info.

so what is the average inflation per month? 10%?

If you opened the spreadsheet it would actually have been the first thing you would see

01/10/2012 – 01/10/2013: 71,41%
01/10/2013 – 01/10/2014: 81,87%
01/10/2014 – 01/10/2015: -6,98%

Okay, so I dont really have the oportunity to reply to every one of you, but I did read all of your comments.

1) A lot have expressed, that you don’t think that Gems should have been included. What I really wanted to show was just what I thought the average player would have brought over a month and how the prices would have changed.
I do think you you have a very valid argument, so I’m going to create another index (along the one I already have) where gems are not included. So I’ll do all the calculations with and without gems to see what the difference between the two indexes are.

2) Other than removing gems I’m going to add some “crafting cost” indexes. So with these indexes I’m going to calculate the cost of crafting full ascended and excots armor, weapons along with some other items. This way we will be able to see what the cost of these things are, since these indexes will not include “luxuary” items like legendaries (I’m doing seperate indexes for every thing).

3) I’ll add a few lower tier materials. This is based on the feedback from Isaih (can’t remember his name). He said that the velocity on these items were actually higher compared to higher tier materials.

4) I’m going to change the layout of the entire spreadsheet, I fell like I’ve a little to many tabs. Still not sure how it’s going to look but I’ve some ideas that I’ll implement over the next week.

5) I know a lot of these changes a based on supply and demand. But as long as we don’t have thoes common items from the real world like food, cloth and rent there is no way to really get past this. We will have to include items like legendaries (I would still argue that they are very important, it’s someting that most players work towards) and crafting materials, and Arenanet are changing recepies and adding new items all the time.

6) Unfortunetly these numbers are actally point based. This is a minor problem, and some trends are therefore lost.

7) Some have sugesting including containers, weapons and other items. As long as there is no where to get these numbers, other than recording it myself, I can’t add that many new items, I simply don’t have time for it.

8) if you have any sugestions for the basket please let me know. Maybe I messed up the weights of some item or you feel like we need to scale it, I would gladly listen to your feedback.

This research might not be perfect, but it still show a genreal changes in prices, whatever it’s deflation or inflation. I hope i covered most things, otherwise let me know.

EDIT: Added point 8

nekretaal.6485:

Nice effort. You have a great platform to set up indecies to model inflation.

(1) High end WvW consumables. Basically foods and oils that serious WvWer pay for.
(4) Set up a index immediately for any Pvp only drops

Any specific items you had in mind. I don’t really play WvW so I don’t know what is consideres “good” amoung the serious players.
What items only drop in Pvp by now, that are not accountbound?

Behellagh.1468:

Gold to Gems shouldn’t be considered in the calculations at all. It’s entirely dictated by the imbalance in gems bought vs gems sold. Gems bought is dictated by gold availability and desirable items on the Gem Shop.

As a good Tyrian, I do put a portion of my gold toward Gems to keep their value up. Even if there’s nothing I want at the moment.

1) Including gems is a grey area; gems are the equivalent of Euros if the Eurozone only produced products that could be sold in the US and Euros could only be spent on those products or converted back into US dollars. I think making an index with and without is the best option. If you are looking for the market wide affect of currency activity, an index without gems will still capture currency effects as it re-enters the marketplace. I am mulling over a better way to capture the effects of it leaving and returning.
5) You are trying to reproduce real world processes and a lot of people are trying to reproduce real world results. A CPI is meant to capture what participants in an economy regularly spend currency on and Tyrians regularly spend gold on gem-shop exclusive items and prestige goods. Stick your your guns about including prestige goods. If real world humans were able to stop buying everything when they turned 18 (lvl 80) but still earned income, the BLS of any rationale country would still collect data on what adults were buying.

9?) The best usable process advice you got was from Ensign concerning price smoothing. Mine was good as well but not really usable since you can never get genuine buying data, your weights will have to be arbitrary. Perhaps players would be willing to collect their purchasing habits for you? Anyway, apply some price smoothing.

Thank you for efforts.

edit:
Price smoothing is extremely important in a marketplace where currency activity is heavily influenced by the demand for non-essential products. Without non-essential products there is no genuine individual demand frequency, only individual demand period. A player can fulfill all their demand anywhere on the history of price points.

Psientist.6437:

1) Including gems is a grey area; gems are the equivalent of Euros if the Eurozone only produced products that could be sold in the US and Euros could only be spent on those products or converted back into US dollars.

The thing is it doesn’t matter because the exchange seeks to balance transactions on both sides, the equivalent of forcing a 0 current account balance. The mechanism through which an exchange would affect the aggregate price level (and thus short run inflation) is through the current account balance, and since that is locked at essentially zero you know its impact on inflation is 0 through an accounting identity.

As far as the gold denominated markets are concerned, the gem exchange is equivalent to one person mailing gold to another (and you can ignore the gem side of the equation entirely); the quantity of gold doesn’t change, just perhaps the categories on which it is spent and some small corresponding velocity effects.

So you really don’t want to combine them; showing them side by side is still valuable however.

Psientist.6437:

Stick your your guns about including prestige goods.

Seconding this. More to the point, you can think of prestige goods as having two major categories – those that are crafted and require a ton of materials, and limited time skins from years past. The former is already being captured by your material index (and if they are not you should include the relevant materials); the latter would be odd to include because there’s no expectation of a regular volume – the price of an old skin goes up as its supply and presumably volume of trade drops. You want things that at least in principle people consume similar amounts of month to month.

Psientist.6437:

Mine was good as well but not really usable since you can never get genuine buying data, your weights will have to be arbitrary. Perhaps players would be willing to collect their purchasing habits for you? Anyway, apply some price smoothing.

I think it would be good to look at what some of the weights in the basket translate into in terms of finished goods though – are the T2-T4 being consumed in proportion with ascended crafting? How about T5 fine materials in proportion with T5 common? The weights aren’t arbitrary, you’re trying to capture real patterns of consumption – so it’d be good to have an underlying model.

Would really re-iterate smoothing though – pull prices from 2-3 days before and after the first of the month and do some weighted averaging over them to calculate a price index. Also, if you’re not doing it already, average over the buy and sell offer prices, as those are going to move together in both of these markets overall while the spread bounces around.

If you’re pulling historical data off of spidy or the like then those old data points should already be smoothed as part of their compression process; just look into how they’re doing their smoothing and adjust your own calculation accordingly.

Really good work and good advice coming in, keep it up.

Okay thank you very much. This is not really anything I’ve done before, I pretty much just read up on a few power point slides from an Economics class I had last year and some information from the ECB, so I’m learning by doing. The econ I have this year is about OCA (optimal currency areas) where we don’t really look into inflation.

I’m looking into some smothening atm. A friend of mine is going to help me code something, so that I can import a lot of data from spidy. This way I should be able to get all numbers from the graphs.

What would be the best way to smoothen the data? Would it be best to take an average over a few days around the 1st of every month. Or maybe even the average of every month.

DennisChrDk.9823

Nice work!
You certainly put a lot of time and effort into this project.

However, I also have a few concerns.
First, You attribute the price increase of the items you’ve put in the “basket” to “inflation”. Which, as you defined as: “a sustained increase in the general price level of goods and services in an economy over a period of time.” But failed to determine the cause of such increase. It seems to be implied that such an increase is due to the increase of gold in the economy. While an increase of gold supply may certainly be a factor, one also cannot ignore the demand on such items, that also increase the price.

With many new players into the game, it stands to reason that as they level up, they are going to want such things. And because there is more demand for them, the price does increase, as they are willing to pay more to obtain them. I would argue that much of the price increase is largely due to demand, and demand alone, rather then more gold in the general economy.

Second, concerning exotic crafting. For many players obtain exotic equipment, especially for players with multiple characters, its far easier, cheaper and faster to get a full set of exotic using karma, pvp reward tracks and dungeon tokens then it is to craft a set. This also effects the prices. That should also be taken into account when considering inflation rates. I am of the opinion that one of the major reasons people are crafting exotics, is to level their crafting discipline in order to get Ascended gear. To which most people either salvage, forge, or sell at a loss in order to recoup the cost of crafting them.

And finally, I don’t believe that gem conversion should be included. It is completely separate to the general economy. The price for gems from gold is only increasing because many people don’t want to spend real money on gems if they don’t have to. You generally start seeing spikes in prices when new items are introduced in the gem store. Which again have no real effect on the general economy. The general consensus of the gem exchange is “Buy gems for gold, it’s easy and cheap. But don’t buy gems for gold because the exchange rate is terrible.” Which only leads to higher prices. With such an attitude, one can reasonably figure that even if the general economy was 100% perfect and stable (that is 0 price increase from launch day), gold to gem prices would still increase.

I actually forgot to update the post, but it should be fixed now.

@pdavis.8031
Some of the thigns you pointed out is actually adressed in the edits. You mentioned that you think that most of the inflation is due to a rise in demand? I agree that it’s a mix of multiple things, but I would still say that some of the reward changes had a very big impact. Something like the changes to dungeon must have have a huge impact and the gold supply would have risen a lot. I don’t think you can just ignore something like this.

quote: Second, concerning exotic crafting. For many players obtain exotic equipment, especially for players with multiple characters, its far easier, cheaper and faster to get a full set of exotic using karma, pvp reward tracks and dungeon tokens then it is to craft a set. This also effects the prices. That should also be taken into account when considering inflation rates.

The other way to obtain excot i actually not something I’ve considered. But I’m not sure I understand the rest of it, about taking this into acount. Do you think I shoud lower oricalcul, gossamer etc. weight in the basket? Because compared to the other items in the basket, tier 6 items is used in so many more recepies, for sigils and a little for ascended armor as well. When you craft from 400-500 you only craft excoti weapons/armors so the amounts you use is much higer.

I’ll do a second calculations of the inflation excluding gems. You are not the first person who pointed this out, and I will change it

DennisChrDk.9823:

The other way to obtain excot i actually not something I’ve considered. But I’m not sure I understand the rest of it, about taking this into acount. Do you think I shoud lower oricalcul, gossamer etc. weight in the basket? Because compared to the other items in the basket, tier 6 items is used in so many more recepies, for sigils and a little for ascended armor as well. When you craft from 400-500 you only craft excoti weapons/armors so the amounts you use is much higer.

This is true. But one should consider the varied uses of those mats. Sure many recipes call for T6, but at the same time, you need to look at how many of those recipes are being used. Take Reaper of Souls for example. It takes 100 vials of powerful blood to make. But how many people are making the Reaper of Souls? According to GW2 spidy, it only has 5 supply, but a demand of 31. This tells me that there is very little demand for the item, but even less people are making it.

So when considering the weight of T6, you should be keeping such things in mind, as that will effect the outcome of the numbers.

DennisChrDk.9823:

I actually forgot to update the post, but it should be fixed now.

@pdavis.8031
Some of the thigns you pointed out is actually adressed in the edits. You mentioned that you think that most of the inflation is due to a rise in demand? I agree that it’s a mix of multiple things, but I would still say that some of the reward changes had a very big impact. Something like the changes to dungeon must have have a huge impact and the gold supply would have risen a lot. I don’t think you can just ignore something like this.

I agree that the dungeon reward changes have increased the gold supply, but with the increased supply, also comes with greater velocity of gold in the system. The way I see it is that people are making a lot of gold running dungeons, but that gold is being spent on the TP in greater amounts, which reduces the amount of gold in the economy, or on gems, which completely removes the gold from the game.

As most people will advocate using the TP over farming for mats, this leads to more people farming gold, which leads to the removal of gold (15% from TP fees) as they buy what they need. Which increases demand, which increases prices. Those people doing the selling will often turn around and buy what they need, which further reduces the gold, and increases demand on other items, which increases prices.

In a very short time, with multiple markets moving so rapidly, it’s not unreasonable to assume that much of the gold made on dungeons, or SW, or other farming activities, is just as quickly being removed through TP fees and other gold sinks.

pdavis.8031:

DennisChrDk.9823:

I actually forgot to update the post, but it should be fixed now.

@pdavis.8031
Some of the thigns you pointed out is actually adressed in the edits. You mentioned that you think that most of the inflation is due to a rise in demand? I agree that it’s a mix of multiple things, but I would still say that some of the reward changes had a very big impact. Something like the changes to dungeon must have have a huge impact and the gold supply would have risen a lot. I don’t think you can just ignore something like this.

I agree that the dungeon reward changes have increased the gold supply, but with the increased supply, also comes with greater velocity of gold in the system. The way I see it is that people are making a lot of gold running dungeons, but that gold is being spent on the TP in greater amounts, which reduces the amount of gold in the economy, or on gems, which completely removes the gold from the game.

As most people will advocate using the TP over farming for mats, this leads to more people farming gold, which leads to the removal of gold (15% from TP fees) as they buy what they need. Which increases demand, which increases prices. Those people doing the selling will often turn around and buy what they need, which further reduces the gold, and increases demand on other items, which increases prices.

In a very short time, with multiple markets moving so rapidly, it’s not unreasonable to assume that much of the gold made on dungeons, or SW, or other farming activities, is just as quickly being removed through TP fees and other gold sinks.

I can’t agree to this. To much speculation on multiple fronts.

Fact 1:

- directly gained gold increases the overall goldsupply in the ingame economy

Fact 2:

- anything farmed to be sold on the trading post reduces the total goldsupply in the economy due to taxation

Speculation 1:

- people who farm dungeons spend more gold on buying things from the trading post

Speculation 2:

- people who sell things on the TP are more likely to buy things as well thus having more gold removed

The only things we can be sure of are the facts. The speculation part is just that, speculation, no matter how probable it might be.

I’d say most people do a mix of things. Run some dungeons, run some Silver Wastes, do some world bosses, do some champ farms, etc. Very few people farm 1 of those things for 6-10 hours straight.

The only thing we can be sure of is inflation du to more gold entering the economy, everything after that is a wild guess.

pdavis.8031:

or on gems, which completely removes the gold from the game.

Careful here, this is untrue. aproximately 28% of that gold leaves the economy. More gold in the gem-gold exchange leads to more gems getting converted to gold. The gold is not removed, it’s merely in a different state and will eventually reenter the game economy. The only gold removed here is the one for the 15% tax both ways.

Cyninja.2954:

I can’t agree to this. To much speculation on multiple fronts.

Fact 1:

- directly gained gold increases the overall goldsupply in the ingame economy

Fact 2:

- anything farmed to be sold on the trading post reduces the total goldsupply in the economy due to taxation

Speculation 1:

- people who farm dungeons spend more gold on buying things from the trading post

Speculation 2:

- people who sell things on the TP are more likely to buy things as well thus having more gold removed

The only things we can be sure of are the facts. The speculation part is just that, speculation, no matter how probable it might be.

I’d say most people do a mix of things. Run some dungeons, run some Silver Wastes, do some world bosses, do some champ farms, etc. Very few people farm 1 of those things for 6-10 hours straight.

The only thing we can be sure of is inflation du to more gold entering the economy, everything after that is a wild guess.

pdavis.8031:

or on gems, which completely removes the gold from the game.

Careful here, this is untrue. aproximately 28% of that gold leaves the economy. More gold in the gem-gold exchange leads to more gems getting converted to gold. The gold is not removed, it’s merely in a different state and will eventually reenter the game economy. The only gold removed here is the one for the 15% tax both ways.

All me to retort.

“Fact 1:

- directly gained gold increases the overall goldsupply in the ingame economy"

I agree with this, and said so in my post. How could directly gain gold not increase gold supply?

“Fact 2:

- anything farmed to be sold on the trading post reduces the total goldsupply in the economy due to taxation"

Also agree. That’s why the TP taxes are a gold sink. It’s designed to remove gold from the game.

“Speculation 1:

- people who farm dungeons spend more gold on buying things from the trading post"

So let me ask this. If people are farming dungeons for gold, and are not spending it on the TP for various skins, mats, etc. what are they doing with it? Are they sitting on it? If they are just storing it in their wallets, then it doesn’t have any impact on the general economy. It only becomes a factor in the economy when it is used to purchase items.
The whole idea behind farming dungeons is that it is faster and easier to obtain the desired items on the TP than counting on RNG for the drop. Therefore, those who farm dungeons for gold, are doing so in order to get gold to buy items off the TP. Either high priced mats, expensive skins, or whatever. If I want that 300g helmet skin, the best and fastest way, without buying gems, would be to farm dungeons and then buy it.

“Speculation 2:

- people who sell things on the TP are more likely to buy things as well thus having more gold removed "

Ok consider TP flippers. They buy one thing, sell it, and buy more things. Also consider the average player. They get a drop, decide they don’t like the skin, don’t need it, etc. and sell it on the TP. They then take that money and put it towards something they do want or need. Of course it may not happen right away, but ultimately they sell and they buy. This is the very nature of the Trading Post. This is a fact. It is not speculation to say that “people buy and sell things on the TP”. Of course they may be a few outliers who refuse to use the TP at all, and prefer to gain everything by farming and gathering, but a vast majority of players will use the TP at least once in their game career.

“Careful here, this is untrue. aproximately 28% of that gold leaves the economy. More gold in the gem-gold exchange leads to more gems getting converted to gold. The gold is not removed, it’s merely in a different state and will eventually reenter the game economy. The only gold removed here is the one for the 15% tax both ways.”

Yes and No. Yes the tax completely removes the gold from the game. Yes the gold used to buy gems is in a different state, but lies outside the game economy. However, considering the the amount of gold it takes to buy gems, and with the general steady increase of that amount, one can easily see that much more gold is being spent on gems, than gems being spent on gold. Of course there are some minor fluctuations here and there. With this trend, and with the gem prices continuing to rise, there is not much of that gold returning into the game. Thus being outside of the game it’s not effecting the economy at all. And so until it does make a return into the game due to gems converted to gold, it is essentially removed from the game, as it’s not effecting anything.

pdavis.8031:

Cyninja.2954:

I can’t agree to this. To much speculation on multiple fronts.

Fact 1:

- directly gained gold increases the overall goldsupply in the ingame economy

Fact 2:

- anything farmed to be sold on the trading post reduces the total goldsupply in the economy due to taxation

Speculation 1:

- people who farm dungeons spend more gold on buying things from the trading post

Speculation 2:

- people who sell things on the TP are more likely to buy things as well thus having more gold removed

The only things we can be sure of are the facts. The speculation part is just that, speculation, no matter how probable it might be.

I’d say most people do a mix of things. Run some dungeons, run some Silver Wastes, do some world bosses, do some champ farms, etc. Very few people farm 1 of those things for 6-10 hours straight.

The only thing we can be sure of is inflation du to more gold entering the economy, everything after that is a wild guess.

pdavis.8031:

or on gems, which completely removes the gold from the game.

Careful here, this is untrue. aproximately 28% of that gold leaves the economy. More gold in the gem-gold exchange leads to more gems getting converted to gold. The gold is not removed, it’s merely in a different state and will eventually reenter the game economy. The only gold removed here is the one for the 15% tax both ways.

All me to retort.

“Fact 1:

- directly gained gold increases the overall goldsupply in the ingame economy"

I agree with this, and said so in my post. How could directly gain gold not increase gold supply?

“Fact 2:

- anything farmed to be sold on the trading post reduces the total goldsupply in the economy due to taxation"

Also agree. That’s why the TP taxes are a gold sink. It’s designed to remove gold from the game.

“Speculation 1:

- people who farm dungeons spend more gold on buying things from the trading post"

So let me ask this. If people are farming dungeons for gold, and are not spending it on the TP for various skins, mats, etc. what are they doing with it? Are they sitting on it? If they are just storing it in their wallets, then it doesn’t have any impact on the general economy. It only becomes a factor in the economy when it is used to purchase items.
The whole idea behind farming dungeons is that it is faster and easier to obtain the desired items on the TP than counting on RNG for the drop. Therefore, those who farm dungeons for gold, are doing so in order to get gold to buy items off the TP. Either high priced mats, expensive skins, or whatever. If I want that 300g helmet skin, the best and fastest way, without buying gems, would be to farm dungeons and then buy it.

“Speculation 2:

- people who sell things on the TP are more likely to buy things as well thus having more gold removed "

Ok consider TP flippers. They buy one thing, sell it, and buy more things. Also consider the average player. They get a drop, decide they don’t like the skin, don’t need it, etc. and sell it on the TP. They then take that money and put it towards something they do want or need. Of course it may not happen right away, but ultimately they sell and they buy. This is the very nature of the Trading Post. This is a fact. It is not speculation to say that “people buy and sell things on the TP”. Of course they may be a few outliers who refuse to use the TP at all, and prefer to gain everything by farming and gathering, but a vast majority of players will use the TP at least once in their game career.

I’ll say it again, no matter how liekly something is, you are guessing as to what people are doing with their gold. Yes, many are “sitting” on their gold until they reach a critical mass and start buying luxury items, which in turn drives lower level item prices.

What i was refering to was to assume dungeon runners would spend more gold on the trading post than other players, which your post suggested. It’s not more likely. I know people who farm dungeons daily and just watch their gold supply increase. Does it affect the economy right this minute? No. But it does when they decide to purchase high value items.

pdavis.8031:

“Careful here, this is untrue. aproximately 28% of that gold leaves the economy. More gold in the gem-gold exchange leads to more gems getting converted to gold. The gold is not removed, it’s merely in a different state and will eventually reenter the game economy. The only gold removed here is the one for the 15% tax both ways.”

Yes and No. Yes the tax completely removes the gold from the game. Yes the gold used to buy gems is in a different state, but lies outside the game economy. However, considering the the amount of gold it takes to buy gems, and with the general steady increase of that amount, one can easily see that much more gold is being spent on gems, than gems being spent on gold. Of course there are some minor fluctuations here and there. With this trend, and with the gem prices continuing to rise, there is not much of that gold returning into the game. Thus being outside of the game it’s not effecting the economy at all. And so until it does make a return into the game due to gems converted to gold, it is essentially removed from the game, as it’s not effecting anything.

Again, no. The only thing this is indicative of is that the ratio of gold to gems is growing more unbalanced. Since gems can be bought with real life money (a major factor of them to be exact) and we have no data on how many gems enter the exchange, all we can say is that the ratio between both is growing. This has absolutely no effect on how much gold is spent on gems or how much gold is entering the game economy.

For example, if suddenly no one bought gems with real life $$$ and converted those to gold, the gold-to-gem price would spike even though no more or less gold was getting converted.

Same goes for if suddenly 3 times as many gems got bought and converted to gold. Even if the amount of gold converted to gems got doubled, prices for gems would still fall.

Cyninja.2954:

pdavis.8031:

“Careful here, this is untrue. aproximately 28% of that gold leaves the economy. More gold in the gem-gold exchange leads to more gems getting converted to gold. The gold is not removed, it’s merely in a different state and will eventually reenter the game economy. The only gold removed here is the one for the 15% tax both ways.”

Yes and No. Yes the tax completely removes the gold from the game. Yes the gold used to buy gems is in a different state, but lies outside the game economy. However, considering the the amount of gold it takes to buy gems, and with the general steady increase of that amount, one can easily see that much more gold is being spent on gems, than gems being spent on gold. Of course there are some minor fluctuations here and there. With this trend, and with the gem prices continuing to rise, there is not much of that gold returning into the game. Thus being outside of the game it’s not effecting the economy at all. And so until it does make a return into the game due to gems converted to gold, it is essentially removed from the game, as it’s not effecting anything.

Again, no. The only thing this is indicative of is that the ratio of gold to gems is growing more unbalanced. Since gems can be bought with real life money (a major factor of them to be exact) and we have no data on how many gems enter the exchange, all we can say is that the ratio between both is growing. This has absolutely no effect on how much gold is spent on gems or how much gold is entering the game economy.

For example, if suddenly no one bought gems with real life $$$ and converted those to gold, the gold-to-gem price would spike even though no more or less gold was getting converted.

Same goes for if suddenly 3 times as many gems got bought and converted to gold. Even if the amount of gold converted to gems got doubled, prices for gems would still fall.

I am thinking we are essentially saying the same thing concerning the gem exchange.

If not, then let me put it this way.
The gem exchange has a pool of gold, and a pool of gems. Every time someone uses gold to buy gems, the price of gems for gold increases as the pool of gold increases, and the gem pool decreases. Each time someone converts gems to gold, gems are added into the gem pool, and gold removed, the price for gems for gold decreases.

Imagine it like a balance scale with gems on one side and gold on the other. If you add gold, and remove gems, whats going to happen to the gem side? It’s going to go up. (indicating increased prices). And if you add gems to the gem side, and remove gold, the gem side goes down (decreasing prices)

Gems bought with real money don’t come out of the gem pool, and have no effect on the ratio of gems to gold, and gold to gems. Therefore, the ONLY factor in determining gem prices on the exchange (both ways) is whether or not people are buying gems with gold, or gold with gems.

Hypothetically, if there were 0 gold sinks in the game at all No TP fees, no WP fees, no vendor prices, etc. and lets say that not a single person ever used the exchange. No one ever bought gems for gold, nor converted gems to gold.

What would happen to the exchange rates as the amount of gold in game increased?
Would the prices of gems for gold go up according to how much gold was in the economy?

pdavis.8031:

I am thinking we are essentially saying the same thing concerning the gem exchange.

If not, then let me put it this way.
The gem exchange has a pool of gold, and a pool of gems. Every time someone uses gold to buy gems, the price of gems for gold increases as the pool of gold increases, and the gem pool decreases. Each time someone converts gems to gold, gems are added into the gem pool, and gold removed, the price for gems for gold decreases.

Imagine it like a balance scale with gems on one side and gold on the other. If you add gold, and remove gems, whats going to happen to the gem side? It’s going to go up. (indicating increased prices). And if you add gems to the gem side, and remove gold, the gem side goes down (decreasing prices)

Gems bought with real money don’t come out of the gem pool, and have no effect on the ratio of gems to gold, and gold to gems. Therefore, the ONLY factor in determining gem prices on the exchange (both ways) is whether or not people are buying gems with gold, or gold with gems.

Hypothetically, if there were 0 gold sinks in the game at all No TP fees, no WP fees, no vendor prices, etc. and lets say that not a single person ever used the exchange. No one ever bought gems for gold, nor converted gems to gold.

What would happen to the exchange rates as the amount of gold in game increased?
Would the prices of gems for gold go up according to how much gold was in the economy?

I fully understand how the gold-gem exchange functions.

What I objected to was you saying that the increase in price of gold to gems is due to not a lot of amout of gold flowing back into the game economy. I’m saying that without knowledge of the increase in the gem pool, this statement can’t be made.

Your hypothetical has nothing to do with the situation. None of the rates would change unless an interaction with the exchange takes place. Higher amounts of gold in the ingame economy would allow for more gold spent on gems though.

Yes, gems from real money have no effect on the exchange, unless someone uses them to exchange them for gold. Since we don’t know how many people do so, we don’t know what the ratio of gems to gold is and why it changes. There are 2 reasons as to why the gem gold price might rise:

A.) people buy gems with ingame gold, thus the gold exchange storage increases and the gem storage decreases. The ingame goldsupply decreases.

B.) people buy gems with real $$$, and use those gems to buy ingame gold. Now the gem storage increases and the gold exchange storage decreases. The ingame goldsupply increases.

If A is severly higher than B, then the price of gold→gems will rise. Since we do not know though how high the influx of gems is, and we can be certain it’s not a 1-to-1 exchange. We can only spekulate as to how much gold flows back into the game economy.

Cyninja.2954:

Careful here, this is untrue. aproximately 28% of that gold leaves the economy.

If John Smith is remotely competent (note: John Smith is exceedingly competent) then the amount of gold leaving the economy through the gem exchange is exactly 0.

pdavis.8031:

But failed to determine the cause of such increase.

Who cares? He’s trying to get an empirical measure of inflation, and is doing a pretty good job of it. Understanding what is contributing to the inflation rate may be interesting but is outside the scope of the project – in fact, without a reasonable empirical measure of inflation there is no empirical basis for any causal inferences on the factors contributing to inflation.

John Smith.4610:

Oh, actually lastly, don’t use precursors, really don’t use anything that doesn’t trade at a reasonable velocity or anything that’s too much of a luxury.

What would be defined as a luxury in this game and why?

John Smith.4610:

Running weights on the goods is something you should do, unfortunately the weighting can be slightly arbitrary and that drastically changes the results, so you end up with results entirely hinged on a series of variables that don’t necessarily have a scientific backing.

Well, it’s not arbitrary in principle – you want to weight things by volume. If people consume a lot of something it has a higher weight than if it’s something people don’t consume a lot of. Of course there isn’t any good public volume data to work with, but that shouldn’t stop you from doing it right internally.

John Smith.4610:

Oh, actually lastly, don’t use precursors, really don’t use anything that doesn’t trade at a reasonable velocity or anything that’s too much of a luxury.

This is interesting. You don’t want to use anything that doesn’t trade at velocity because the price data is low quality, and thus introduces noise into your measure and makes it unreliable.

You don’t include luxury goods for a different reason though – you leave them out because people don’t buy luxury goods simply because they are higher quality, but because they have reputation externalities – people buy things that are conspicuously expensive because they are expensive, not despite it, in order to show off their wealth. How much people choose to spend on reputation externalities is interesting unto itself, and is certainly capturing something of economic interest, but it’s definitely not an end-around way of measuring changes in the ruler that we call money, so to speak.

I don’t really know what that means in an economy like the GW2 one, though, where reputation externalities are a driving force of the economy. Thinking about it is actually raising more questions about its exclusion from standard measures than answering why it shouldn’t be included. Thoughts?

John Smith.4610:

Neat stuff some things to keep in mind is in a game economy players control then printing of currency irl governments tightly control that.

Inflation was a useful construct before the 1940s. People could measure it just fine when the basis of exchange was the quantity of shiny metal dug out of remote holes in the ground and transported for storage in a different hole in the ground.

John Smith.4610:

Also demand on items change greatly based on updates which can look like inflation but is just a shift in demand (this is more clearly seen in the precursors, or silk) this makes it hard to judge purely on an index alone.

If you have a price surge in a given sector like that, with no real inflation there should be a counterbalancing drop in a different sector; if some prices rise without others dropping, individuals have lost buying power and you have prima facie inflation. It’s an aggregate measure because you are trying to say something about the money, not about individual commodities.

John Smith.4610:

Reposting Izzy’s and my reponses from before:
John Smith:
This is interesting, but I would point out some things to look into if you try this again. As Izzy said, we make a lot of balance changes so you have to keep major changes from balance changes in mind. Running weights on the goods is something you should do, unfortunately the weighting can be slightly arbitrary and that drastically changes the results, so you end up with results entirely hinged on a series of variables that don’t necessarily have a scientific backing. Lastly, be very careful of putting variables in a model that feed into each other, that will cause the model to over-represent changes in a single set. Arguably you could fix this by proper weighting, but this gets really tricky when there’s a lot of variables mixing together. Oh, actually lastly, don’t use precursors, really don’t use anything that doesn’t trade at a reasonable velocity or anything that’s too much of a luxury.
Overall though, I think this is a great effort given the resources you have to work with and a really interesting read.

Izzy:
Neat stuff some things to keep in mind is in a game economy players control then printing of currency irl governments tightly control that.
Also demand on items change greatly based on updates which can look like inflation but is just a shift in demand (this is more clearly seen in the precursors, or silk) this makes it hard to judge purely on an index alone.
We spend a lot of Econ balancing on managing supply and demand of each material and currency and tweak both sides of that equation. For example butter was rock bottom at the start but we rebalanced its in/outs so it’s increase is not from inflation.
Just some other ways to think about this stuff

Every player i know takes worth on gold, so you using ectos as an index is a throw-back to the original, when people had to use ectos since the system didn’t support platinum wealth.

This is likely why ectos are so stable and gold is, well, not so much. KISS

Ensign.2189:

pdavis.8031:

But failed to determine the cause of such increase.

Who cares? He’s trying to get an empirical measure of inflation, and is doing a pretty good job of it. Understanding what is contributing to the inflation rate may be interesting but is outside the scope of the project – in fact, without a reasonable empirical measure of inflation there is no empirical basis for any causal inferences on the factors contributing to inflation.

I disagree. Spending habits certainly play a factor in tracking inflation. If certain markets are increasing in price, finding the cause of that price increase is important. Instead, if you simply show that prices are increasing, without examining the “why”, all you have is a list, or a chart, that shows prices are going up. At that point it’s easy to say “See!? Prices are going up because more gold is entering the economy, and not enough is leaving. This is classic inflation!”
It then becomes a theory and an opinion, and not anything more.

Thank you for entering the conversation.

John Smith.4610:

Reposting Izzy’s and my reponses from before:
John Smith:
This is interesting, but I would point out some things to look into if you try this again. As Izzy said, we make a lot of balance changes so you have to keep major changes from balance changes in mind. Running weights on the goods is something you should do, unfortunately the weighting can be slightly arbitrary and that drastically changes the results, so you end up with results entirely hinged on a series of variables that don’t necessarily have a scientific backing. Lastly, be very careful of putting variables in a model that feed into each other, that will cause the model to over-represent changes in a single set. Arguably you could fix this by proper weighting, but this gets really tricky when there’s a lot of variables mixing together. Oh, actually lastly, don’t use precursors, really don’t use anything that doesn’t trade at a reasonable velocity or anything that’s too much of a luxury.

Overall though, I think this is a great effort given the resources you have to work with and a really interesting read.

If we had access to this balancing data and usage data, I am sure we would only use it for the greater good. I am using the royal we, my opinions of what others would use it for has no scientific backing.
Your two ‘lastlies’ seem connected; precursors are made from T5-T6 materials, rare and exotic weapons. A lot of the currency activity information generated by precursors is also present in and could be captured from the information generated by those materials and weapons. However if the velocity of Precursors is low enough, their sale shouldn’t generate too much noise. That noise may be worth it since the price of precursors includes a fee for service in the form of a risk aversion premium. Hmmm…maybe with proper weights even that service good be measured in T5-T6, rare and exotic weapons. All things considered, the argument for or against including precursors could only last until HoT goes live.
Are you willing to answer a question that would help Dennis?

“Anything that’s [not] too much a luxury” or stuff that propagates currency activity at relative velocities:

Would you include ascended insignias and ascended crafting materials? These items would also capture information about the market value of services such as collecting account bound ascended materials, earning laurels, and leveling crafting disciplines.

Ensign.2189:

Cyninja.2954:

Careful here, this is untrue. aproximately 28% of that gold leaves the economy.

If John Smith is remotely competent (note: John Smith is exceedingly competent) then the amount of gold leaving the economy through the gem exchange is exactly 0.

No, absolutely no. I’m not even sure how you come up with this idea.

The gold – gem exchange has a built in goldsink feature. It’s one of it’s major strengths. It has been covered multiple times that the total goldremoval from the game comes out to be about 28% (15% both ways.). If you were to buy gems from gold you lose 15% value. If you then convert those gems to gold again you lose another 15% (in total about 28%).

Why you would want to keep this at 0 I have no clue. The goldinflux to the game economy through players is certainly high enough that this goldsink is absolutely needed.