EFTA Document EFTA01780732

From: Joi Ito < Sent: Saturday, April 23, 2016 4:39 PM To: Jeffrey Epstein Subject: Re: rough draft of a paper on accounting OK > On Apr 23, 2016, at 12:35 PM, jeffrey E. <jeevacation@gmail.com> =rote: > some good points.. " in search of certainty," could be its =ubtitle. . I suggest more drilling on " value " . on types of hacks =o current system. etc > On Sat, Apr 23, 2016 at 12:12 PM, Joi Ito < > wrote: > Super rough draft on something I'm thinking about writing =bout accounting and book keeping. > Tell me if you think it's worth writing and if you have any =dea or > suggestions... > — > Reinventing Bookkeeping and Accounting > Double-entry bookkeeping was deployed in its modern form in the 13005 =nd while minor innovations have occurred since then, the fundamental =tomic unit of tracking and managing value - our accounting system - is =till based on this 700 year old invention. With modern computers, =etworks and cryptography, we have the opportunity to fundamentally =hange one of the most important and empowering, but also limiting =lements of modern civilization. > Accounting underlies finance, business, the management of the =tate's resources and is the way that the world keeps track of =lmost everything of value. While companies will keep track of how many =idgets, contractual obligations, employees they have and many other =hings, the accounting system and the laws that support it, require us =o convert just about everything into monetary value and enter it into =ome ledger system using the 700 year old double-entry bookkeeping =ethod. > When you take, for instance, a contract that pays out a million =ollars if it rains tomorrow, and put it into your accounts, you will be =equired to guess the chance of rain - maybe 50% - and value that asset =t something like $500,000. The contract will actually never pay out =500 - it will either be worth zero or a million dollars in the end, but =f you were forced to trade it today, you'd probably sell it for =omething close to $500,000 so for tax and management purposes, you =E2 value" the contract at $500,000. > A company's accounts are full of cells with numbers in them =hat represent a sort of numerical value denominated in some currency - =en, Dollars, Euros - and those numbers are added up and organized into = balance sheet and a profit and loss statement that shows the health of =he company to investors and profits to calculate the amount of tax owed =o the government. >' > At the very top level, your balance sheet has a list of assets and =iabilities. If you looked in the assets column you'd have a =unch of things that you would be reporting as having value including =hings like printing presses, lines of code, intellectual property, =bligations from people who may or may not pay you in the future, cash =n various countries' currencies and bets on things like the =rices of a commodity in the future or the value of another company in =he future. EFTA_R1_00099475 EFTA01780732 > As an auditor, investor or a trading partner, you might want to drill =own and try to test the assumptions that the company is making or test =hat might happen if some of the assumptions that the company were =aking changed. You might also want to understand how buying the company =ould change your own company based on the way your obligations and bets =nteracted with theirs. Today you would rack up millions of dollars in =uditor fees to "get to the bottom" of most big =ompanies. The process would involved manually reviewing the contracts =nd the assumptions made in every cell of every spreadsheet. That'= because accounting is a very "lossy" process reducing =omplex functions with probabilities and dependencies into static =umbers at every step. The underlying information is stored somewhere, =ut most of it requires manually digging around. > The modern financial system is full of companies who have figured out =ays to guess when investors and the companies themselves have made =istakes in their assumptions. These companies bet against the companies =nd financial assets with inaccurate pricing or are somehow able to take =dvantage of the gap in information and convert this into financial =eturns. Also, when these mistakes are duplicated across the system it =ause fluctuation amplification which also allows companies to make more =oney both as markets rise as well as fall if they can guess in advance =f those fluctuations. In fact, as long as the whole system doesn'= collapse, smart traders make more money on fluctuation than on =tability. > Just like rodent exterminators aren't excited about the idea =f rodents being completely eliminated - they would not longer have jobs = those financial institutions that make money by "making the =ystem more efficient and eliminating waste" don't =eally want a system that isn't wasteful and is very stable. > Right now, the financial system is built on top of a way of thinking =bout money and value that was designed back when all we had was pen and =aper and where reducing the complexity of all of the web of =ependencies and obligations was the only way to make the system =unctionally efficient. The way to reduce to complexity was to add it up =nd simplify it. The current technology just builds on these 700 year =ld building blocks trying to make the system "better" =y doing very sophisticated analysis of the patterns and information =ithout addressing the underlying problem of a lossy and over-simplified =iew of the world. This view of the world being that everything of =E2 value" should be as quickly as possible reduced to some =E2 value" where "value" is a number =enominated in "money." > Today, we have the technology and the computational power to create a =ystem of accounts that isn't as lossy and in fact could retain =nd deal with a lot of the complexity that the current system of =ccounts was meant to avoid. > There is no reason that every entry in our books needs to be a number. =he cells could be an algorithmic representations of the obligations and =ependencies that it represents. In fact, using machine learning, =ccounts could become sophisticated probabilistic models on what might =appen depending on how things around it changed with the "value =80 of any system being different depending on who was asking, where =hat person was, and in what time frame. > Today, when the Financial Stability Board conducts a stress test, it =ives a bank a scenario - changes in the credit markets or the prices of =ertain things. The bank is then required to return a report on whether =t would crash or remain solvent. This requires lots of human labor and =ork to go through the accounts and run simulations. What if the =ccounts were all algorithmic and you could instantly provide the answer =o the question. What if you had a learning model that could answer a =ore important question - "What sets of changes to the market =OULD make you crash and why?" That's really what we =ant to know. We want to know this not just for one bank, but the whole =ystem of banks, investors, and everything that is interacting =inancially. > When I'm buying something from a company - lets say a credit =efault swap from your company, AIG. What I want to know is whether, =hen the day came to pay the obligation on the impossibly unlikely =hance that the AA mortgage backed bonds that I was betting against =efaulted, you'd be able to pay. Right now, there is no easy way =o do this. However, what if all of the obligations and contracts, =nstead of being written on paper and recorded as numbers, were actually =omputable and "visible." You'd immediately be =ble to see that, in fact, in the scenario in which you'd have =o 2 EFTA_R1_00099476 EFTA01780733 pay me, you'd actually have no money since you'd =ritten similar contracts to so many people that you'd be broke. right now, even the banks themselves can't see this unless an =nternal investigator figures out to look for it and finds it. > With systems like Enigma, there are ways that we might be able to =vent keep these accounts open to each other without compromising =usiness and personal secrecy and privacy. While computing every =ontract as a cell in a huge set of accounts every time anyone asked a =uestion would exceed even today's computing capacity, but with =achine learning and the creation of models, we might be able to dampen =f not stabilize the massive amplifications of fluctuations that occur =oday because we are building our whole system on a house of =ver-simplified cards with the handlers having an incentive to make them =ragile and opaque - to introduce inefficiencies that they can exploit =ater to make money. > I think that the current excitement about Bitcoin and the Blockchain =ave created a great opportunity to rethink the fundamental system of =ccounts. I'm much more interested in this than apps for banks =r even new ideas in finance, which will address some of the symptoms =ithout taking a shot at eliminating one of the root causes of the =mpossibly complex and outdated system that we've build on a =lever trick invented by traders of the 1300s. > -- > please note > The information contained in this communication is confidential, may > be attorney-client privileged, may constitute inside information, and > is intended only for the use of the addressee. It is the property of > JEE Unauthorized use, disclosure or copying of this communication or > any part thereof is strictly prohibited and may be unlawful. If you > have received this communication in error, please notify us > immediately by return e-mail or by e-mail to jeevacation@gmail.com, > and destroy this communication and all copies thereof, including all > attachments. copyright -all rights reserved <?xml version=.0" encoding=TF-8"?> <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> <plist version=.0"> <dict> <key>date-last-viewed</key> <integer>0</integer> <key>date-received</key> <integer>1461429520</integer> <key>flags</key> <integer>8590195713</integer> <kepgmail-label-ids</key> <array> <integer>22</integer> <integer>2</integer> </array> <key>remote-id</key> <string>606441</string> </dict> </plist> 3 EFTA_R1_00099477 EFTA01780734