welcome to the beanpod a podcast about decentralized finance and the beanstalk protocol i'm your host rex before we get started we always want to remind everyone that on this podcast we are very optimistic about decentralized finance in general and beanstalk in particular with that being said three things first always do your own research before you invest in anything especially what we talk about here on the show second while you're doing that research try to find as many well-developed opposing viewpoints as possible to get the best overall picture and third never ever invest money that you can't afford to lose or at least be without for a while and with that on with the show [Music] on this episode of the pod we're going to be talking about soil on previous episodes we've discussed beans stock and seeds and pods the last of which is the dead instrument that beanstalk uses to make its credit based system work but some listeners out there may wonder how beanstalk decides how many pods to offer to potential borrowers during a given season the answer my friends is soil so how soil works its importance to the protocol's debt level and interest rate and a number of other related topics are what i'll be talking about with bubles today publicly it's great to have you as always on the podcast welcome back thanks for having us back rex indeed so today we want to talk a little bit about soil now soil is kind of a fundamental component of beanstalk and really for the purpose of refresher how about you just start by running us through what soil is and why it's important absolutely so soil is the willingness to borrow beans and in particular it's the willingness for beanstalk to borrow beans so if we think about what it means to be a credit-based stablecoin beans are the the assumption that a user is making of beans on their underlying stability is fundamentally one of the credit of beanstalk as a protocol and what that means is that beanstalk anytime the price of a bean is too low tries to meaning there's excess supply of beans in the market beanstalk tries to borrow beans from the market in order to remove enough beans from the market to return the price to a dollar so soil is the willingness to borrow beans and it is a very important part of beanstalk because ultimately when the price is below one it is soil issuance that serves a major role in helping beanstalk return the price to the peg so right along with that there have been some changes to how soil works within beanstalk over its over its lifetime um would you want to walk us through just some of those changes and how those improvements have affected how the protocol acts sure so soil is once again the willingness to borrow beans from the market now the question is at any given time how many beans is beanstalk willing to borrow from the market how much soil is there what is the soil supply and so the amount of soil available is one of the things that has been improved over time so originally beanstalk issued soil that was cumulative over time from season to season up to a maximum and the concept was that if there was multiple consecutive seasons where there was an excess of beans in the pool call it ten thousand beans uh over the previous two seasons on average then beanstalk would issue would have outstanding twenty thousand soil or have issued twenty thousand soil assuming no one had lent to it yet there would be twenty thousand soil so the concept is soil would accumulate over time to reflect that the fact that uh the cumulative total excess of beans in the pool uh that was what the soil was originally designed to reflect and in particular the there were a couple problems with that model the first problem was that the soil itself would grow the amount of soil that was available would accumulate significantly such that when beanstalk would have an extended period of time where the price was below a dollar the amount of soil would almost always accumulate to the maximum which was defined as a percentage of the bean supply and so the the concept is that when the bean price was below a dollar for an extended period of time there would be a buildup of soil and that had a couple of problems with it the first problem was if we go back to one of the core principles of beanstalk which is that it needs an efficient market for soil it needs to incentivize lenders to lend to the protocol in an efficient fashion the fact that soil would accumulate over seasons made it such that let's say there's an excess of 10 000 beans this season the if i'm a lender that wants to lend to beanstalk but i want to lend 20 or 30 or 50 000 beans to be in stock i would actually be inclined to wait to lend to beanstalk once there's 10 000 soil to wait for the soil to accumulate potentially and while at the margin the first in first out harvest schedule would still have an effect on people's willingness to borrow particularly when you had an extended period of time where no one was lending to the protocol this buildup of soil was uh not helpful to the efficiency of the soil market in general now the other problem or problems i should say with the soil as it was originally defined was that the amount of soil would accumulate to a maximum as a percentage of the bean supply and there was also a minimum soil available as a percentage of the bean supply so and we'll i assume we'll talk about it in a second when it comes to measurement of demand for soil and why beanstalk measures demand for soil but the concept is beanstalk requires uh the ability to measure demand for soil every season and so it needs there to be some amount of soil outstanding every season and originally the minimum soil supply every season was a function of the total bean supply i believe it was 0.1 percent of the total bean supply and it didn't factor in the weather and the problem with that was when you would be in a period of time where there was excess demand for soil the protocol was actually issuing a ton of debt that it didn't need to be issuing and so the original soil model had a couple of problems with it the accumulation of soil over time an arbitrary maximum soil that would build up too and then a minimum soil that prevented uh beanstalk from deleveraging effectively whenever there was excess demand for soil so those were the the problems if you will that were uh quite evident uh in the original in the original model and since then there has been a a number of changes to the to the system so the let's talk about what's actually changed the first is that the soil no longer builds up from season to season such that if there's a 10 000 bean excess in the pool over the previous season and nothing changes over the next season there will continue to be only 10 000 soil available beanstalk doesn't issue cumulative soil and therefore there's no need for a maximum instead bean stock only issues every season the uh time and liquidity weighted average excess of beans in the pool over the previous season and it's designed to just on a single season trailing basis to try to remove all of the excess beans from the supply from the market the the other change is with regards to the minimum so there's no longer a minimum as a percentage of the bean supply but instead when the price is above one and again there's no cumulative so now you really only have two scenarios you have the price below one and then the soil is a function of the excess or you have the price above one in which case there's a shortage of beans and therefore under i believe it was bip nine uh that could be wrong uh that introduced the improvements to the to the soil supply when there's excess demand for soil such that the the current amount for soil when the price is above one is equivalent to uh basically answers the question of how many pods were paid back at the start of this season the amount of soil available at the start of this season is the based on the weather the amount of soil such that the amount of pods issued during this season will be the same as the pods paid back at the start of the season such that the pod line will stay a fixed rate or a fixed length assuming that uh the pod line will stay a fixed length assuming that the there's excess demand for soil and the price is above one so that's a little history of how soil has evolved over time and what it currently looks like there's probably some further improvements that can be made for example adding some sort of coefficient to the pods minted at the start of the season such that the pod length pod line length can actually decrease as opposed to staying fixed that would be a marginal future improvement that could be made and hopefully will be made uh prior to being stuck restarting or shortly thereafter and otherwise that's basically the current state of how soil is issued i appreciate that and i feel like some of this also could go off into a separate discussion that maybe we should have in another time about the length of pod line and some of the feedback that the the protocol has gotten about the length of the pod line over time but that's definitely a podcast for uh for another day um so you did talk about demand for soil and obviously this is a really important component of of how soil is issued a lot of complexity in here would you want to talk us through that demand measurement process sure so the issuance of soil is distinct and independent from the demand for soil but from a peg maintenance perspective measuring demand for soil is is very important so at a high level what beanstalk tries to do is to gauge whether the soil that it has issued the beans that it is willing to borrow whether there's demand for that soil whether people are willing to lend to beanstalk and people's willingness to lend to beanstalk can be a function of an infinite number of things but the only thing that beanstalk really can change substantively to make it more attractive to lend to the protocol or less attractive to lend to the protocol is to change the interest rate is to change the weather and so fundamentally what beanstalk does is at the start of every season beanstalk increases or decreases the weather by up to three percent based on the price the debt level and the change in demand for soil over the previous two seasons so the price and the debt level are a little bit simpler uh where the price is either above and below a dollar the debt level is either excessively low reasonably low uh reasonably high or excessively high uh because again the this the system how it how it should react and how it should adjust the interest rate is different depending on how indebted the system is at the moment how aggressive should it be as a function of how indebted it is at the moment so the third factor that plays into the change in the weather every season is the change in demand for soil over the previous two seasons so fundamentally why is measuring demand for soil important it's because beanstalk needs to know whether it should increase or decrease the interest rate for lending to beanstalk and in practice other than the price and the debt level the other main factor that from a theoretical perspective it makes sense to consider when adjusting the interest rate is purely the market demand for for the debt and so what beanstalk does beanstalk is designed to be as simple as humanly possible or as simple as mechanically possible we might say so the goal is to classify demand for soil not uh not in the most complex way possible but simply to classify it either as increasing steady or decreasing and i mean there's probably an argument to be made for just having an increasing and decreasing demand for soil uh but the there is some some something nice about having a third class with which to choose from and at the end of the day as we'll get into how measurement of demand for soil is measured it's pretty easy to see how you could do without the steady state if you don't want to but generally uh demand for soil is classified as either decreasing steady or increasing based on the change in demand over the previous two seasons so we can get into how it actually works in practice but what beanstalk is doing is trying to figure out over last season and the season before it how has the demand for soil changed and if demand for soil is increasing how should bean stock respond as compared to if demand for soil is decreasing so how's the demand for soil measured well it's actually a non-trivial problem because at the end of the day you you want the protocol to be able to measure soil in a way that is not easily gameable so what that means in practice is there's a couple of cases that stand out right one is all of the soil has been sown in a season so beanstalk issues soil and all of that soil is sown the the concept is that should very clearly be a case right all of the soil is sewn but as how do you define all the soil is zone is it if there's x soil is it that you need x soil to be sown uh or what if there's 0.01 soil left is that all of the soil to be sewn so there's there's some complexity around what it means for there to be all all of the soil to be sewn what beanstalk does is it says if there's anything less than one soil left that counts as all of it so basically the way beanstalk measures demand for soil is either all of the soil has been sown meaning there's at almost one soil left or there was some amount of soil sown over the previous season some some unit of soil and basically what beanstalk does is first it compares the the amount of soil that was actually sewn now there's there's a lot of cases that the demand for soil actually handles so the first thing is just whether or not the amount of soil uh that was sown over the previous two seasons uh what the ratio is between the soil sown last season divided by the soil shown two seasons ago and the based on the rate of the amount of soil sown over the previous two seasons demand can be classified as increasing steady or decreasing and i believe it's if it's greater than 105 percent uh it's increasing demand and if it's less than 95 percent it's decreasing demand and anything in the middle is steady but as we were talking about earlier if the beans are sown in all of the soil or all but one of the soil in a season that is sort of a special case so if all of the soil was not sown last season but all the soil was sown this season that's clearly increasing demand it used to be that it wasn't selling out now it's selling out that's increasing demand now if you have consecutive seasons where soil is selling out then what beanstalk does is it uses the time that it took for all of the soil to sell out to determine whether the demand for soil is increasing or decreasing so it uses amount it uses whether or not all of it was sown and then assuming all of it was sown in consecutive seasons it uses the time and in doing so beanstalk is able to determine uh generally pretty well although still not perfectly whether demand for soil is increasing steady or decreasing and if we go back to the this rate right 105 and 95 it would be very trivial to make the rate one and say anything above one is increasing and anything below one is decreasing but the current model does include a steady state which remains to be seen whether it in the grand scheme of things that will be helpful but uh it was designed to allow for the model to have more flexibility than less so that that kind of leads us into another good part of this discussion let's start with a scenario where where demand is increasing how does that play out in terms of a how how the soil supply itself reacts does that does that necessarily influence it and really b how does that affect the weather and then subsequent potential actions by participants so the the soil the amount of soil that is issued is i guess you can't say totally independent of the demand for soil because in theory if people demand soil they may need to buy the beans from the market and thereby change the excess of beans in the pool but in general the soil supply issuance is an independent calculation from the measurement of demand for soil now if there is an increase in demand for soil over the previous two seasons that affects the the protocol's response when it comes to changing the interest rate so the the unfortunately there isn't really a short answer uh in the sense that when the protocol it basically analyzes three axes so you have price which has two cases you have debt level which has four cases and you have demand for soil which has three cases so two times four times three is 24. so you have 24 cases based on the price the debt level and the change in demand for soil and what beanstalk does is it has a effectively a table that it executes off of whereby based on the price the debt level and the change in demand for soil bean stock increases or decreases the interest rate the weather based on the table so the there's a couple nice things about this one is that the actual contents of the table can be adjusted quite easily via a bip so the the measurement of the the state of the system is one thing but then the response to the state of the system is sort of a separate thing so the the way that being stock takes the measurements is very generalized but the actual responses can be tuned or fine-tuned over time so how beanstalk responds when the demand for uh soil increases or decreases is largely a function of the the other two factors the price and the debt level as well so it's not a it's not necessarily a one-to-one uh thing per se that's a that's an important point and i guess what i would say to listeners is there's a lot of detail on this in the white paper itself and so published just talked through the that kind of multitude of different cases those cases are spelled out with a lot of clarity in the white paper itself so i'd direct anybody that wants to learn a little bit more and read those in detail rather than just having publius you know sit and read pages out of the white paper i i think the the only thing like at a high level that can be said though is when there is an increase in demand for soil when people are more willing to lend to the protocol bean stock tends to be based on the the figures in the table beanstalk tends to be more aggressive in its responses when it comes to the interest rate meaning beanstalk is willing to lower the interest rate more if there's an increase in demand for uh soil when the price is too high or it will raise the weather less if the price is too low and there's increasing demand for soil so in both cases the change in the interest rate is gonna if there is an increase in demand for soil that's likely to make the weather net net lower than it otherwise would be if there was a decrease in demand for soil so that's the intuition very fair point so the whenever we talk about soil one of the other things that always comes up as a topic especially immediately leading up to the exploit was how do participants get a hold of soil you know the with soil splice being relatively low at that point there's a lot of discussion about things like soil sniping bots and what alternatives there might be um that the protocol might offer and uh do you want to uh you want to talk through um you want to talk through that process sure so the the state beanstalk wants to be in is where there is high demand for soil if we go back to the stability assumption of bean users the bean price will be stable at a dollar if there is demand for soil so bean stock it doesn't really care what interest rate it pays as much as it cares that there is demand for soil so and it would rather overpay to clear all of the soil regularly so for the month and a half maybe prior to the attack maybe more than that actually i think it started in february uh so february march april two and a half months prior to the attack all of the soil that was available at the start of each season was being sewn into so people were lending the maximum and it was highly competitive and the only winners were bots uh that were all competing and there were lots of failed so transactions and in fact uh there was a bip to change the way that the sow function works to minimize the amount of failed so transactions because of how competitive it was but the the short answer is if you can't access the soil because there's excess demand for soil uh beanstalk there's lots of other ways to participate in the system uh including buying pods on the pod marketplace which is the secondary market so even if you can't lend to beanstalk directly and get pods you can buy pods on the secondary market uh or you can participate in the silo going forward you'd also in theory be able to buy fertilizer so lots of different uh one of the one of the principles of beanstalk is that you need a diverse set of uh competitive uh competitive participants in the market that all have different risk and return profiles and strategies and therefore that's where you get a competitive market from so at the end of the day lots of different ways to play this and soil is designed to be the most attractive such that everyone would prefer soil and beanstalk again maximizes for demand for soil or optimizes for it so publius just mentioned the opportunity to buy fertilizer for those of that of you out there listening um starting on monday june 6th the barn raise is going to officially begin which is our opportunity to recapitalize beanstalk post-exploit and one of the main functions of that barn rays will include the opportunity to purchase fertilizer and we'll have other information about that on the podcast and there's a ton of really good information that's been published by beanstalk farms encourage anybody to go to our twitter feed or to the bean.money site where there's information available about how to get involved in the barn rays publish anything else you wanted to you wanted to cover on the topic any other things that you think we should we should talk through well the the long and the short of it is that soil is a a useful tool to understand the state of the system the willingness to borrow is a little bit hard to to quantify right so soil is a nice abstraction of that and then understanding demand for soil demand for willingness to borrow beans it's a little weird so it's demand to lend to the protocol i think soil has been a nice within the context of the farming metaphors the soil has been one of the best pieces of the metaphor to help explain things but obviously there's a lot of nooks and crannies so it's very helpful here to walk through it all absolutely and i definitely appreciate it um publish as always thanks for your time thanks for joining us on the podcast thank you sir