Building A Decentralized Consumer Price Index (CPI): A Paradigm Shift in Economic Transparency

Transcript

[Music] presenting the next speaker Brenda Loya uh the stage is yours thank you all right so my name is Brenda I am the CEO and one of the co-founders of teller and today I'm not going to be really really talking about teller I'm going to be talking about how we need to expand in from we need to go from just using crypto as um to remove intermediaries to expand our Horizon to

basically start doing things that we can um do better in the from the traditional system that does not just involve making money and this talk it specifically talks about the CPI I think it's um one of the most important economic measures that affect everybody broadly um this quote is by Ben Franklin he was one of the founding fathers of the US and initially he says taxes right we canot avoid taxes but as you all know if you have enough money especially in the US you probably can um but inflation it doesn't matter if you're rich if you're poor where

wherever you are you face is inflation and inflation is basically just um measures the change of the average cost to buy a basket of goods and services over time so that's all that it does it's very simple measure uh but it's used really broadly by governments to do all sorts of things that directly have an effect on you um they adjust government assistance programs like um uh in the US for example we have this uh food stamps we have uh payments to retirees um it also use to evaluate whether or not you qualify for this type

of help and as well as cost of living adjustments so in general if you work in a city like New York you're probably going to earn a higher wage than if you you know do the same job in El Paso Texas where the cost of living living is much lower and that cost of living adjustment is generally based on inflation and just the how much it cost to live in the city and also if you are in a traditional or even if you work for the government mostly I don't see this as much on crypto but your wages on a yearly basis are generally adjusted by this inflation measure just to keep you basically being able to afford what you

are able to afford right now so you can afford it next year and then obviously mon AR policy the US uses this inflation measure to directly impact the money supply in in circulation the money supply in circulation is really important because it's basically a way for the government to and I was talking to somebody earlier and I think it was pretty Innovative so I'm stealing it um to basically devaluate your money or take money away from you on a on a yearly basis so if inflation increases by 2% each year each year your purchasing power decreases by that much and that's just accepted right so this is a measure that comes out

every month that not a lot of people realize how of and how directly it affects their pockets and how important it is to them with that in mind let me tell you a little bit of how the CPI started the CPI started because Republicans and Democrats wanted to know who was right about implementing the Tariff Act of 1890 basically they increased the you know these tariffs really high and like was like 50% like something you know we had a surplus and they wanted to basically decrease the amount of imports that we were getting so they implemented this and then they

were like let's see who's right and who's wrong and use this next time uh when we run for you know Congress or whatever and that's how it started super politicized and then it didn't take and then the very first um survey only included white guys married guys they lived in the city that's it that's that that was the the extent of the survey that's who they wanted to know how they got impacted and that was how they were going to drive their their policies and it did it took 50 years for them to actually start including other groups like Hispanics and blacks in the US so 50 years of this inflation measure and doing policies based on this that

completely excluded full portions of the population and then it took another after that they included them but they also included them very purposefully under represented and it took another 40 years for them to actually go back and do a revamp of survey where instead of just collecting specific items in in a basket or you know bread egg butter whatever now they changed it to actually do uh categories so they brought in the survey they sort of try to tackle a little bit of uh the bias that was already there but it's still a an imperfect measure to change methodology for the CPI it's almost like trying to move a mountain

it's very difficult because it can be very politicized and a lot of research goes into that the CPI has a lot of criticisms starting from substitution bias um lacked inclusion of of new items poorly measures qu um changes in quality these are all things that we have known from the beginning of the survey there's a lot of people that focus on this and I I want to tell you substitution bias is just because I saw a post on Twitter three days ago talking about how the price of or of meat went up from 10 to $20 but yet uh the government is assuming that you're just instead of eating meat you're just going to eat

some burger that only went up 10 cents and so they were disagreeing with the CPI because they were like it's under measuring and I'm not just as well off before because I wanted steak and it's way more expensive but I don't want to substitute it for a burger and that's basically what substitution bias is it's like basically it's assumed that you are still keeping your standard of living you just shift from eating pasta to eating more rice or whatever else it is that you eat more often because it's cheaper and so you're still at the same standard of living but you substitute so that's that's what that means um for new items basically the basket of goods or the things that are tracked over time by the government it's um it's identified at the beginning of the survey and then

it's tracked over time so over time obviously some of those things become obsolete so maybe they were tracking like iPhones 14s and now we're on 15 or 16 whatever number they are not now um they're not equally um if they're and they're not included right they're they're different things they're not included or maybe we start we all start using some other platform to do something Innovative and that's not going to be included because he hasn't been there for for it to be tracked and quality changes is basically I'm sorry going back to the um the iPhone like if you're if you're trying to track prices for the iPhone 14 and now it's the 15 you have to make an adjustment for me the camera is better so it doesn't show

up as just inflation but the true or the the elephant in the in the room I think it's basically the conflict of interest of what the government is providing as a measure of inflation and how they actually use it so the government actually gives their their employees um an adjustment each year based on on the inflation measure so if I'm going to give everybody an increase and I am reporting a 3% CPI then everybody in inflation index then everybody gets a 3% increase but if I end up calcul in a 20% one uh then everybody gets a 20% so there is a conflict of interest there right not just there but also in their political agendas depending on what's

happening in the country and I think in the US I think they're doing okay I mean we we try our best we try our best but like there are some examples um that are really interesting to me like the one in Argentina Argentina has not been underw they have resources they have ports and yet from 20 2007 to 2015 the government openly admitted to just cooking the inflation numbers they just made them up basically um sort of um and so they did this because obviously they had political agendas they wanted some um loans from the uh World Bank and you know so if if your inflation is really

high then you're your money is not worth that much and it might be more difficult for you to be able to pay your debts so you want to make sure that you have a lower inflation rate and that's what they did you know they instead of reporting the whole inflation rate they reported about a third of it so Venezuela I don't even know what's going on there but U it's also one of those cases where if you cannot trust your government because they're corrupt how can you trust the the statistics that they're providing you Zimbabwe is another example where uh I think what they did is they decided to use old methodology because they for for an implementation of of um of one of their policies because they didn't like the new one and that's just you know what

the government decided to do so these are just a few examples of what's going on in the world right now um I think Argentina right now is doing a little better their inflation rate is being um reported more accurately right now but they're like at 300% or something crazy like that they'll eventually I think uh come down to a normal CPI with their new policies hopefully but in general we have to to think about this um government statistics are basically provided just by the government and we just assume or we're required I guess to just trust them right they are obviously looking out for us but in the US we have antitrust laws

because we want to protect our consumers when there's no competition when when there is no oversight one there is no no reason for the company to want to do better to want to improve their products and so if the government has no oversight has really you know and and it's the end all be all of these statistics and these are the official statistics that are used by all of these legislators what is their drive to make it better right and so um I don't think there is one but now we have technology that can allow us to actually create something that is independent and provides this oversight so what I'm proposing is basically a structure where

if you want to do any type of research you publicize it on a ipfs or something like that that it's a decentralized storage platform and then obviously you do have to find incentives to start that um development and then but once you do all the raw data needs to be also saved on a dspo or ipfs or something similar the reasoning behind that is because right now we have the methodology available for what the government says they are doing with the data however we don't have the underlying data to actually go and research it and make sure that they apply the methodology the way that they say that they applied it so the next

step is also publish your methodology where anybody and everybody can find it and apply it and make sure that you're doing what you say you're doing and then um publish the outcome of your research also on chain and then this what this creates an open peer review everything is out there and if you don't agree with my methodology or whoever else's methodology go look at the RW data make it better publish your own methodology and do it again like that that you know you don't have to redo the whole cycle but if you do think that in general like the actual survey was flawed there's a whole [Music] um you know a a whole class a whole uh

studies that go into just how you word questions so that you don't bias people's um people's answers so if you think that it was flawed then redo it again and start over but then once you make it transparent you allow everybody to make sure or to to do that peer review for you for the CPI um what I'm proposing is using methodology that has already been used in the past Alberto cavalo and Manuel bertotto I believe they Argentina Argentinian uh but they started this project called the a billion prices project in MIT and they went back to Argentina and

tried to recreate their CPI numbers from 2007 to 2015 because there was that Gap when it was basically uh made up or and so they went back and recreated it using scrape data they went back they did all the research they were able to this is before 2020 so back then so when they did this in 2017 when they recreated the data they were able to see that the majority of the prices online were fairly correlated with the prices offline so they went back they recreated the the whole um the whole period and what they ended up coming back with is

after they went through everything was that basically the way that the numbers were being uh manipulated were not at the B at the micro dat level micro dat is like the actual pricing information it's very difficult to get a whole room of people to agree on politics yet alone like try and figure out what all liars to to include or not include especially if they're not in the same party or whatever different ideologies but what they were doing or what they think they were doing based on what they came up with they were basically just they had this the correct Trend they were just dividing it by a third so everything was being processed fairly correct and then what it was what was actually publicized

for the people was just divide it by a third and call it a day and that's uh and anyway with that in mind using their methodology I think uh it should be more than good enough now that after the pandemic a lot of offchain prices you know have on have gone online at least in the US the majority of our shopping I think people just moved or changed their behavior so instead of shopping on on actual stores we do a lot of shopping online and so that provides a wealth of information and data that we can use now that it's going to be in my you know even more correlated than that's a theory because I haven't seen the data but if they were able to observe that before 20120 I think we should be able

to observe similar Trend now uh the biggest thing for uh um for any type of survey data it's allr detection and the government when our taxes what they basically pay for is for us to for for them to have all of these people looking at at this data and cleaning it up and classifying it and then just um processing it through an aggreg you know through an aggregation software so that you guys so that we can all see an index so the majority of the money goes in all detection and classification so if we can automate those to be more efficient we can do this way cheaper and uh produce something of Fairly good quality

um to be able to provide this measure that should be complimentary to what the government is creating right if if they have nothing to hide if they're doing everything the best that they can do with the information that they have then it should be very very similar the problem arises when it actually starts challenging the government statistics so with that in mind that's one of the reasons that everything should be saved on a DSP or an ipfs or something like that and you have to use an oracle that is censorship resistant right we want to make sure that even if uh the government says like shut down that the protocol still works and that's why like I'm I'm going to be using teller obviously um censorship

resistance there's no fee for new uh for new uh data feeds so that's really important because it's a this is a common good and eventually it should be owned by the community so anybody should be able to tip it and be able to create and incentivize a CPI for a specific geograph Geographic um area there anybody can dispute it anybody can stake um they uh so teller is going to have its own layer one so it's going to make it really easy to just Bridge the data over and then once you have this U measure of economic measure on chain that is openly available to everybody and it becomes robust enough then you can actually use it to create stable

coins and create other things on top of it and that is completely independent from the government especially in areas that where you can't really trust the government um so the way that it works is very simple it's really three contracts uh one contract where you register your your research your methodology and the outcome and then um another contract where where it's going to be different contracts for each type of research that you're doing and you know you just collect the data and you track the reputation of the people of the data scrapers mostly and then teller is just um it's just an oracle it could be another Oracle too um I'm focusing on them because teller is censorship

resistant and I think that's necessary in this case so in they would be just used to aggregate data which is aggregate the data and just report back the CPI on chain and then track the provider reputation which is just basically something that we would um Define how to do and then they just aggregate the information and provided um we have currently um everything that we need to be able to create this and that's what I'll be working on now like I said I do run teller but I think that we have been focused for far too long on just bringing offchain data on chain we can actually do some of these we can create

better data using our systems and and not depending on archaic architectures that were just there because really that was the the best thing that we could do at that moment in time so and obviously we don't want to be hunted by people's political agendas uh I know this is a public good and it might seem like it's sort of a dream but if you think about it the US government at least for now they collect about 880,000 prices of different items each month and they collect this for free nobody pays for that data um and in 1942 I'm trying to see am I still

good yes with time okay so in 1942 the US government implemented price controls for different types of of food like meat bread the war was going on and what they did they actually took a bunch of uh mostly Housewives CU they usually were the ones going out and shopping for for groceries and they trained them they would give them each week a set of prices for each item and they were actually the enforcers they did this for free so people do a lot of things for free for the government already and they did that

I mean they they no longer longer have this in the US but they were able to implement this policy this US policy wide um thing with just volunteers on you know on foot obviously there's a benefit to them because they you know if if the a store was increasing the price over what was the the control price that was provided for them they had to pay less right so they they were incentivized because they knew like okay this is not the price you have to bring it down and then they went back and and reported to the government um so this can absolutely be done it has to be Community funded and it has to be Community Driven and it

has to be geographically um focused because it has to be the community that wants to drive this and push this forward what I'm going to do and what I'm I'm trying to build right now is building the the very first one and I think the one of the safest plac and also because I live there I feel like the US is fine you know it's like it's where I know it's it's what I know and I think there shouldn't be too much of um resistance even if it comes out hopefully because I do think that they are doing the best thing that they can do with the data that they have so that is what I am building right now but I'm happy to answer any questions in terms

of methodology or like um random s sorry yeah thanks I wanted to ask so you want to work on that in us right why did you decide to have a speech in Prague so what why did you decide to have the speech in Prague in Czech Republic so I this is going to be Community Driven it's going to start in the US because it's easy for me that's where I live I live it yes but yeah um I think that the majority of the benefits for something like this are going to be in um I don't know if it would be in Prague but definitely Latin America and

I thought that this would be a great place to talk about it I don't know if how much um how much people trust or don't trust their governments in this side uh the world but basically I just it wasn't really like I'm going to go to prag and present this I came to dcon and presented this I'm going to be here being in Belgrade um I do want to make um make my way back to Latin America I know that that's I'm sorry I'm looking at the thing it's keeps it keeps moving um that's where it has been needed in the past but the three examples that I provided I mean Zimbabwe was somebody that somebody somebody else uh mentioned

it to me and I was like huh you know it's not just the and and I'm sure there's plenty of people in the US that also don't trust the government and don't trust that the CPI is correct but like if you have if you if you really disagree then join and let's try and build this and let's you know let's provide some oversight but yeah I didn't specifically say like Prague people I didn't think prag people don't trust their government so yeah you you just mentioned that uh you this should be area specific and it should be Community Driven and is it more like you're seeking for some Community here or just sharing it across the world and checking if if this problem exists also in

different countries if people talk think about that yeah I would like to know like because um it's easier to create an index for for an area right when you because you can collect more information so you need uh you cannot create an index from like three prices you have to have Community or you know data scraping depending on how you're going to build it um so the two things that I want to do is the CPI which would be uh doing data scraping with a bunch of data so that would be like um but then the Big Mac index is really is something else that already exists and it's mostly collected just um from different McDonald's around the world like you want to make sure that you collect all

the McDonald's in one area all the McDonald's in other area to be able to to create that type of index that makes sense yeah that's it thank you so if you guys don't trust your governments just call me sorry is there any other questions thank you very much um um I guess that like for example in prag we kind of trust the government statistics but because we have no other choice like this is the change of the Paradigm right um my question to you is um you mentioned that people might be doing this stuff for free because they did it in the past also the

tell um um and I'm wondering like this does not seem to be sustainable well so teller has no fee for new new price feeds you do eventually So eventually you do have to tip or incentivize reporters to go get the data but like there are some oracles that will charge you thousands and thousands of dollars just to support one new feed and Teller doesn't as long as you provide the definition so there is some yeah so if eventually that's what it's a public good so just like we use taxes to pay for the government to have all of these people doing classification and Loud lier cleaning up the data we

have to have something similar but it that you know it can go the way that I ambition it is going to be a dow and um and then you know you just tip tell her directly so that way if we go away and that has already been specified and the infrastructure is there it can continue to survive without me sure yeah that makes sense thank you yeah any other questions hello uh I wanted to ask uh the Consumer Price Index is dependent on the prices which can be scraped and you have the data but it is also dependent on how you set up the basket of the

goods and I think this choice is always arbit array so how how it should be done in this case yeah so you have to look at the data there's going to be it's going to be very similar to the CATE it's going to follow the same categories that the US has the same Big C categories um and just yeah I mean it's so it's not just that it's also the weights of the category so initially because we don't have data already out there aside from what the government is providing it would have to lean on that so you would get the the weights for those categories and then as you scrape the data and you classify them into those categories that's how they they would be weighted

for the at least I you know when you get started once you have enough data you can actually come up with like better better or more independent weights on how people are using this then we can uh we can shift away from what they provide right but I think initially um especially based on their research that's what they did and and it worked okay it doesn't mean that we wouldn't eventually move to a more decentralized and more independent way of of doing it but yes that's always hard I mean how you how you ask the questions what kind of items you include um that's that's that's always or or you

can collect right because we're data scraping so whatever we can get our hands in it's going to be like as long as it falls in that category and there's a previous price for it then we can actually use it and for example if you choose some basket and I would disagree with that and I will start the this dispute how this will be resolved um you can do your own methodology right so if if you the raw data that's why the underlying data is going to be on the on a DSP so if you don't agree with the way that I classified each of the things then you can do it your own way the reason it's not EX a basket of goods anymore that is just like bread eggs and

potatoes it's because you have like for example I'm Mexican we eat a lot of corn right so in the US if you're just measuring bread and potatoes and eggs you're not capturing my inflation rate which I guarantee you I'm going to buy tortillas this you know Friday uh so so that's why now it's more of a basket it's it's a classification of goods so as long as it's food it's thrown into the food um classification and that's how it's tracked sorry anyways that's all the time we have thank so much Brenda for giving that a great presentation and I hope you have a great time at eth prog thank you thank you so

much thank you for mcing for me