On the Matter of Why Bitcoin Matters

Marc Andreessen was a big part of turning the Web into a mainstream experience, but seems to misunderstand Bitcoin profoundly, writes Glenn Fleishman.

Update: Marc Andreessen has been ridiculously gracious on Twitter, responding to questions my article raised posed to him by other people on the social network, and directly to me. I’d suggest reviewing his Twitter feed as a good adjunct to reading this critique.

Marc Andreessen wrote an essay for the New York Times about Bitcoin, “Why Bitcoin Matters,” in which he attempts to explain the relevancy of the digital currency for the future of commercial transactions. He uses analogies, allegories, history, and ostensible facts to build his case.

However, I believe he fundamentally misrepresents or misunderstands key aspects of the technology, ecosystem, and impact, despite Andreessen Horowitz, of which he is a founding partner, having just under $50m in investment (fully disclosed) in “Bitcoin-related startups.” I own no Bitcoins; Marc has a “de minimis” amount. (I will note that someone owning Bitcoin investments and not Bitcoins is the same as owning gold-mine investments and no gold.)

I recently spent weeks researching an article about the technology challenges affecting Bitcoin for the Economist: “Bitcoin under Pressure.” Some of my insight is drawn from talking to miners, academics, and software engineers about Bitcoin’s past, present, and future.

Pained analogies to the past

Andreessen attempts to draw a parallel between the birth of the personal-computing era and the explosion of the publicly accessible Internet and Bitcoin. These are fundamentally flawed analogies.

A mysterious new technology emerges, seemingly out of nowhere, but actually the result of two decades of intense research and development by nearly anonymous researchers.…What technology am I talking about? Personal computers in 1975, the Internet in 1993, and – I believe – Bitcoin in 2014.

There’s a significant difference between anonymous and unheralded or perhaps even so many millions of people we don’t track them individually. While there are a number of significant milestones and breakthroughs in both PCs and the Internet on the road to mass adoption, there is no single figure like Satoshi Nakamato, the pseudonymous creator of the Bitcoin protocol and initial software implementation. (He may be an individual or a collective that operated under that name.) We know where CPUs, MS-DOS, TCP/IP, SSL, and the like came from and sometimes even every step of their design. (Marc was involved in creating SSL, for instance.)

Personal computers were the result of continuous innovation and change in which part of the developments remained proprietary (such as chip-making technology), and part was fully exposed to the light of day in academic research and shipping products. (Read John Markoff’s excellent book, What the Dormouse Said: How the Sixties Counterculture Shaped the Personal Computer Industry, for some insight into the cultural aspects.)

The Internet grew as a result of well-documented government funding and with the initial participation of many thousands of people in academia, government, and private industry working together. Much of this was public, and I could list off hundreds of people involved in the creation of the Internet, and the popularization of the Web—including Andreessen!

Andreessen wants to paint parallels to show how Bitcoin has a fundamentally similar origin story, when there is little in common. By trying to normalize Bitcoin against previous technologies—PCs being primarily hardware with a significant software component and the Internet primarily data with a significant hardware component—he wants a general audience to accept that an entirely philosophical and intangible new thing is the same as these familiar old things.

Trust me, even though you don’t know me

I agree with Andreessen that Bitcoin is the first practical, large-scale mechanism to deal with the problem of decentralizing trust—no parties need know each other nor trust each other for transactions to complete successfully, verifiably, and irrevocably.

Bitcoin has a remarkable set of interlocking mechanisms that make it exceedingly hard for any individual or group to accumulate enough power to distort the public block chain, a sequence in which one set of transactions builds upon the next. If such distortion took place, it would become known, and various scenarios have been painted about what kind of response would occur. (Some suggest Bitcoin would collapse, and thus that no party would engage in attempting it, however impractical it would be to achieve.)

The scale of computation required at the moment for Bitcoin’s operations, discussed in my Economist article, is so large that only a government-scale effort would be able to overcome the system’s checks and balances. These operations are run by what you could consider as volunteers, called miners, who receive a reward in Bitcoins for their efforts as part of the system’s basic infrastructure. (Bitcoin’s worldwide computational output is currently nearing 200 exaflops—200,000 petaflops—or 800 times the combined capacity of the top 500 supercomputers in the world.)

I also agree completely with Andreessen that Bitcoin can be used for an enormous number of non-currency related purposes in which permanent, irreversible proofs of transactions are required. Bitcoin’s procedure of building a chain, in which each link is cryptographically related to the link before it, means that after an hour or so, it should be impossible even with a world superpower’s efforts to reverse out a transaction.

Further, these transactions can be proven to have been created by a party who possesses a private key; no other party can create such a proof. (The key can be stolen, thus changing ownership, but that stands outside of Bitcoin’s reliability.) One of the fundamental current software architects of Bitcoin, Mike Hearn, explained to me how such transactions could be used to validate identities for discussion forums and for passports.

But here’s where Andreessen gets squishy. He wants to define Bitcoin as both a currency and a “new kind of payment system.” The former is a matter of debate; the latter is provably true today, whether it persists or not.

As a payment system, he asserts Bitcoin has enormous advantages over current credit-card and bank-processing networks.

Bitcoin is the first Internetwide payment system where transactions either happen with no fees or very low fees (down to fractions of pennies). Existing payment systems charge fees of about 2 to 3 percent – and that’s in the developed world. In lots of other places, there either are no modern payment systems or the rates are significantly higher.

That contains a bunch of problematic statements. Without getting into the weeds, many Bitcoin transactions can be carried out without fees or very low fees, but transactions are only committed into the global block chain when miners accept those transactions: miners create links in the chain that are built of transactions. The minimum rate to pay, when a fee is paid, is based on the size of the transaction in bytes, and starts at 0.0001 BTC. The average is currently about 20¢ at the current exchange rate, even if the transaction involves millions of Bitcoins. One can pay more, however.

Transactions with higher fees attached are committed more quickly. As the rate of transactions has increased, the fee has seemingly become more important, as no-fee and low-fee transactions can linger for longer periods without being built into a block and made permanent and thus useful for consummating a deal. This is an ongoing concern, and Bitcoin’s software team has a plan to offer more flexibility with how fees are set. (The team has no specific power; it has to convince a supermajority of participants in the Bitcoin ecosystem to adopt software updates.)

Andreessen masks several upcoming problems with his seemingly straightforward statement:

• Bitcoin is somewhat illiquid and highly volatile, and the fees for moving Bitcoins in and out of legal tender, like dollars, can be in the several percentage point range. It’s not always easy or possible to convert from Bitcoin to cash, and the volatility means the potential loss (or gain) of dollars before the exchange closes.
• Bitcoin miners receive an award of 25 coins each time they succeed at a computational problem that allows them to add a permanent link to the chain. That’s $25,000 every eight minutes or so right now at the current exchange rate. However, the reward drops in half every four years or so. In mid-2016, it will drop to 12.5 coins. The total number of coins that will ever be made is fixed.
• The system gives up these coins, diluting the pool of money at a fixed rate, to miners, which represents a kind of hidden fee, like issuing new shares in a company and giving them to employees. The speculation and volatility currently hides the effects of coin creation as a dilution.
• As the production of coins drops, mining fees are expected to pick up the slack. While they will almost certainly not rise to the current rate used in credit-card transactions, they will likely be non-trivial to keep miners rich in incentives to operate their computer gear (which is currently hard to keep profitable because of the rapid increase in computational power across the network).

Andreessen then moves into advantages related to fraud:

…there are no chargebacks – this is the part that is literally like cash – if you have the money or the asset, you can pay with it; if you don’t, you can’t. This is brand new. This has never existed in digital form before.

But that is only true within the Bitcoin system. As far as can be determined, transactions are irreversible after a short period of time and cannot be counterfeited. The system doesn’t prevent theft or misuse.

Chargebacks occur in the credit-card processing world, as do stopped checks, because of a dispute between a buyer and seller. Credit-card companies and banks mediate in disputes as a backstop to keep buyers happy and because of oceans of government regulations. On the other side, sellers by percentage of sales volume tend to be multi-billion-dollar companies, who otherwise have a lot of leverage against consumers and most other businesses.

When Andreessen says there are no chargebacks this is the point of view of a merchant or seller who deals with fraud. He writes,

…people can trade with Bitcoin (anywhere, everywhere, with no fraud and no or very low fees)…

The seller can’t be defrauded. The buyer can.

Because a Bitcoin transaction can’t be reversed, it means that the party transferring value has no recourse within the Bitcoin system to reverse or dispute a transaction. (Hearn told me there’s actually a mechanism for using third-party dispute-mediation process in the Bitcoin specification, but it’s not implemented in a way to which users have access.)

Thus, someone who wants their money back has to go to court, and companies that accept Bitcoin as payment aren’t above the law. This will move chargebacks from intermediary mediated settlements to small-claims court and higher civil courts.

Bitcoin doesn’t eliminate fraudulent transactions; it only eliminates counterfeit payments. This can, of course, save many tens or hundreds of billions of dollars a year globally and translate to more efficiency in commerce. But removing the intermediary also removes recourse outside of courts, and the cost and nature of that can’t be determined.

Because Bitcoin is like untraceable cash, the process for solving a dispute would likely follow the same rules as for cash. When a transaction occurs over the Internet, the odds of recovering one’s Bitcoins when a seller fails to meet his or her obligations is the same as if you’d sent a wad of bills in an envelope through the mail.

It also glosses over theft. Because Bitcoin relies entirely on the private retention of secrets (private keys that prove ownership of given Bitcoins), a stolen Bitcoin is only traceable within the system to a certain extent, and thefts of millions and tens of millions of dollars have already occurred. Because committed transactions are irreversible, stolen Bitcoins are valuable and nearly laundered.

The discussion of fees and fraud thus also excludes the overhead to businesses. Companies will have to deal with fraud under the laws of the government under which they operate. If a firm processes a $10m order paid for with Bitcoins, and then it is determined the Bitcoins were stolen—what happens then?

…many online merchants are forced to turn away 5 to 10 percent of incoming orders that they could take without fear if the customers were paying with Bitcoin, where such fraud would not be possible.

It’s perfectly possible; it just doesn’t happen at the payment level. It happens a “layer up” in the societal compact in which commerce occurs.

We already have laws in America (and most countries) about receiving counterfeit currency (it is valueless to the recipient when discovered) and about fraudulent transactions: the wronged party can clawback money from whatever parties accepted it, even if the parties accepted it without knowing it was stolen. (This is partly how the Madoff trustee has reclaimed lost billions from those who received payouts.)

Cash, check, or charge?

Critics of Bitcoin point to limited usage by ordinary consumers and merchants, but that same criticism was leveled against PCs and the Internet at the same stage.

The issues for PCs and the Internet were about utility, not about whether or not they actually existed and had value. You could buy a PC; you could get an Internet connection. The cost and complexity of both started out so high, that early adoption was low, but the curve of acceleration for uptake was extremely high.

With Bitcoin, we already have hard cash and credit cards that can be used for digital transactions, and we can transfer money as bits. Bitcoin is a replacement or supplement for an existing creaky and somewhat broken system that nonetheless works.

So Andreessen pivots. Is it a currency or a payment system? If value is retained and used entirely within the system, it’s a form of currency. But he writes:

Bitcoin can be used entirely as a payment system; merchants do not need to hold any Bitcoin currency or be exposed to Bitcoin volatility at any time. Any consumer or merchant can trade in and out of Bitcoin and other currencies any time they want.

That’s currently not true at the scale at which Andreessen envisions this working. However, greater adoption could switch the verity of this statement and allow low-fee (but some fee) Bitcoin exchanges that take place in short periods of time. It’s simply not the case today. Much of his argument, including instant transactions at Target (your Bitcoin account is instantly filled if needed; Target instantly exchanges Bitcoins back to dollars), relies on such a future.

On another front, Andreessen overstates how readily trackable Bitcoins are when discussing theft and illegal activities:

Bitcoin is pseudonymous, not anonymous. Further, every transaction in the Bitcoin network is tracked and logged forever in the Bitcoin blockchain, or permanent record, available for all to see. As a result, Bitcoin is considerably easier for law enforcement to trace than cash, gold or diamonds.

This is only the case because of the current implementation. I wrote about a paper delivered in October that showed how readily transactions could be “unpeeled” and tracked to given addresses, but not to individuals. However, the paper’s authors note that relatively minor changes in the Bitcoin software libraries would dramatically reduce such tracking. It’s almost a mistake that tracking today is possible.

Further, there are proposals for fully untraceable alternative currencies. Zerocoin, initially a proposal for inserting such capability inside Bitcoin, will launch as a freestanding currency. Using zero-knowledge proofs, people will be able to push coins into Zerocoin’s blockchain and take coins out later without any mathematical ability for outside parties to connect the transactions. It’s a laundering service, without any negativity meant. It breaks the chain of knowledge.

Finally, Andreessen moves to the problem of “unbanked” and “underbanked” individuals: the huge number of people in the world who have no access to conventional banking or only limited access and face high fees for use. But he seems to ignore the current existence of mobile payments, such as M-Pesa, which provide Bitcoin-like benefits in an existing ecosystem. Bitcoin has to beat mobile payments for ubiquity, ease, and control in such markets.

(He covers a lot more in this essay, which I either lack enough domain-specific knowledge to discuss, or which is more related to its use as a currency.)

The surprising part to me in this essay is the conclusion:

Far from a mere libertarian fairy tale or a simple Silicon Valley exercise in hype, Bitcoin offers a sweeping vista of opportunity to reimagine how the financial system can and should work in the Internet era, and a catalyst to reshape that system in ways that are more powerful for individuals and businesses alike.

On that, we both agree. Bitcoin shows a path for massively more secure, reliable, and sensible ways to store value and move it around. As a currency, I have little faith that it will become a replacement for dollars, euros, or renminbi. As a model for a future payment and transaction system, I believe it’s already shown its value.

Glenn Fleishman is the editor and publisher of The Magazine, and a frequent contributor to the Economist. He hosts the podcast, The New Disruptors. He’s been on the Internet and its predecessors since 1986, and founded one of the first Web hosting companies in 1994—thanks in part to Marc’s Netscape.

This article was produced by The Magazine. It costs $1.99 per month for two issues or $19.99 per year for 26, and subscriptions include free access to over 150 past articles — our full archive. We commission original reported articles and essays, and run five in each issue every two weeks. You can get a free, seven-day trial via our iOS app or our Web site to try us out.