Twitter. Facebook. AirBnB. Marc Andreessen, co-founder of the$4.2 billion venture capital firm Andreessen Horowitz, has backedthem all—along with dozens of others. His latest project? Upendingfinance. Bloomberg Markets magazine interviewed Andreessen at thefirm's headquarters in Menlo Park, California. Here are histhoughts on the future of finance:

Out with the Old, in with the New

“We have a chance to rebuild the system. Financial transactionsare just numbers; it's just information. You shouldn't need 100,000people and prime Manhattan real estate and giant data centers fullof mainframe computers from the 1970s to give you the ability to doan online payment.

''You would not today, starting from scratch, invent any ofthese financial businesses in the same way. To me, it's all aboutunbundling the banks. There are regulatory arbitrage opportunitiesevery step of the way. If the regulators are going to regulatebanks, then you'll have nonbank entities that spring up to do thethings that banks can't do. Bank regulation tends to backfire, andof late that means consumer lending is getting unbundled.

“We're not going to go backward. When people start doingthings a better way, it kind of doesn't matter what the old waywas. You can find people who will say that this is all just anarbitrage on the current trouble in the financial system, and I'msure the big traditional banks will fight back and try to getthings outlawed.

''But think about the scenario of a loan officer talking to aprospective client. To software people, that looks like voodoo. Theidea that you can sit across the table from somebody and get a readon their character is just nonsense.

''Lots of industries are changing in a similar way. There's beena qualitative approach, and now there's a quantitative approach.Everybody who grew up in the qualitative approach hates thequantitative approach and considers it a giant threat.”

Big Data

“There is a growing idea in Silicon Valley that there aresources of data on consumer behavior we can use to predictcreditworthiness. These will be completely different than thetraditional approach to credit ratings, which are tremendouslyimprecise and 'laggy.' PayPal can do a real-time credit score inmilliseconds, based on your EBay purchase history—and it turns outthat's a better source of information than the stuff used togenerate your FICO score.

''The hypothesis is that there are many other similar sources ofconsumer data: credit card bills, social-network behavior,potentially even search history. Lots of people, both in the bigInternet companies and at startups, are trying to get at theselarge pools of data and figure out new ways to do scoring. Whatthey all have in common is that they are all being done outside ofbanks.

''The minute any of these new credit vehicles can show any levelof repeatability and reliability, the hedge funds come in andprovide the funding. Hedge funds are very comfortable with analyticmodels. If you have sufficient stability, you can getleverage.''

Cryptocurrencies

''The startups chasing disruptive technology aren't workingwithin the existing system. This is the cryptocurrency phenomenon.If it works, we can re-implement the entire financial system as adistributed system, as opposed to a centralized system. We canreinvent the entire thing.

''Bitcoin is clearly in this category. With bitcoin, there are advantages to decentralizing the financialsystem in order to do commerce. For instance, you can make paymentsin all but four countries. So, on day one, you can use it all overthe world. Forget about all the different currencies and bankingsystems; it's a truly universal way to transfer value.

''That also means we have the chance to radically lower fees.Most consumer transactions are weighted with a 3 percent fee;remittances run up to 10 percent, which I think is a moral crime.There's a big opportunity to take those fees out.

''Bitcoin is like technology that's arrived from Mars, and soregulators don't know what to do with it. That's a good thing. Whata lot of financial technology entrepreneurs will tell you is thatif you're going to innovate in financial services, you want to dosomething so new and so different that the existing regulatorysystem doesn't know how to react to you. That is your window ofopportunity.

''The problem with building a new product or service in theexisting financial industry is that tens of thousands of pages oflegislation and thousands of lobbyists are going to come down onyou very quickly.

''We needed a new technology to have the wedge to be able toenter the market, to be able to justify all the work to rebuild thesystem. With bitcoin, we now think we have that wedge.

''Bitcoin is a classic venture capital endeavor: It will eitherwork or it won't. And if it doesn't work, we will lose all ourmoney. But if it does work, it will work in a spectacular way. Ourinvestments will pay off 1,000-to-1 or 10,000-to-1 or some othercrazy extreme, because these markets are so big.'

''There are serious conversations going on at every single largefinancial institution. They're all trying to figure it out. Thetechnology folks who work for the banks are very smart, and thetrading floors have split between 'This is for real' and 'This is ajoke.' Yet people are leaving those institutions to start bitcoincompanies. We're seeing hundreds of bitcoin entrepreneurs. Some ofthem are brand new to financial services. They don't know whatthey're in for, but they're excited.”

Volatility Will Vanish

“Bitcoin had a chicken-and-egg problem. Speculation was the wayto get the thing established. The speculators are doing us all afavor by valuing it at $500 or $600 or $800 because otherwise itwould be valued at zero. Now you can actually do things with it.With scale and time, the volatility will vanish. The designer ofthe system, I think, would say that the system is workingperfectly.

''If you're used to living in a world where you trust people andinstitutions, then it probably does seem weird and bizarre that youwould trust something where you don't know who created it. But ifyou're in a world where you trust numbers and math and code, itdoesn't matter. There is no appeal to authority. You evaluate it onthe basis of the math. It's just math; it's completely open. Noneof us thinks we need to know who created it.

''As for security concerns, bitcoin is designed to work properlyin an untrusted, networked environment. Take Target. As a consumer,there is nothing you could do about Target. Target blew it; theirsystems stunk.

''In a bitcoin world, things like the Target hack are notpossible. The way a digital currency works is that it lines up theincentives to protect yourself with the consequences of failing toprotect yourself. Bitcoin is a digital-bearer instrument: If youhave the numbers on the coins, you own the coins. You can makepayments without having to give any information about yourself, andeveryone can double-check their transactions. If someone hackedinto Target, they would be able to steal all of Target's money—butthey wouldn't be able to steal your money.

''For five years, many of the world's best mathematicians andcomputer scientists have been studying bitcoin and trying to figureout what's wrong with it. They haven't found anything yet. Everycritique people have of bitcoin, so far, can either be answeredwith 'the designer anticipated it and has a solution built into thesystem' or 'there's a service that can be built on top to addressthe problem.' That's the magic of why everyone out here is soexcited about it.”

Copyright 2018 Bloomberg. All rightsreserved. This material may not be published, broadcast, rewritten,or redistributed.