Joi Ito's Web

Joi Ito's conversation with the living web.

Goodbye letter from Lehman Brothers

I get a sense that this sudden "wait a sec, do we really need investment banks anymore?" is the tip of the iceberg of a collapsing of a whole genre of financial services and products that are just obsolete once you have easy access to information and low transaction costs.

While derivatives and securitization of risk so other people can buy it makes sense, the incentive of a company full of people getting paid a crapload of money to do this is to make products that they can sell for more than they're worth. Even with regulation, risk information is omitted and the focus is on selling, not on providing the buyer/user with complete information and low transaction costs. Instead, information is hidden and there is transaction friction everywhere from management fees and transaction fees.

Hand made markets like this made sense when transaction costs costs were high and all this stuff needed to be aggregated and managed, but it doesn't REALLY make that much sense anymore.

This is similar to academic publishers who charge professors to print their papers in journals that the publishers then charge schools and libraries to read. What the professor wants is to be most widely read. Before the Internet, this made sense and was worth the hefty ($10K+ for a single physics journal subscription) price because having physical distribution to libraries was the only way to make sure your paper was distributed to every school. With the Internet, it just doesn't make sense anymore. Now the copyrighted materials can't be viewed without subscriptions and a huge percentage of our academic works are "dark" to developing nations and non-academics.

I think there are a lot of industries and services which were once super important, but have yet to face the reality of our low-cost digitally connected world.

My guess is that even in financial services, there is a bunch more stuff to go. As I bounce from bank to bank trying to find the cheapest exchange rate, I just can't imagine why things like foreign exchange just can't happen with less friction and more information. Banks want to charge a ridiculous percentage/spread/transaction fee and I have so much incentive and ability to avoid this. In fact, I bet MOST things banks do could be redesigned to happen without them.

I can imagine a sort of Web 2.0 for financial services with paypal-like services that store your money and all kinds of financial transactions provided by a huge number of global providers. This would drive competition to increase quality, lower price and would also foster an industry of analysts and information providers who independently helped you make the right decisions. If all of this could be integrated, much like how a lot of the consumer Internet services are grouping up around open standards and specifications, you could get rid of a huge chunk of costs and skewed interests where the people who are giving you advice are also the people who are trying to make money off of your transactions.

Of course we need regulation and fraud protection, but I'm sure there is a more effective way to do this than having some bureaucracy oversee a bunch of overpaid suits who don't understand technology.

My apology in advance to my friends who work at banks. I don't think banks will go away, but I think they need to aggressively adapt to technical changes and rethink why they exist and what people need them for and stop relying on legacy notions and government regulation to protect them from competition. As we have just recently witnessed, things can change VERY quickly.

I'm not a banker or an economist, so what do I know... but just some thoughts from an amateur.


Interesting thoughts - but I think finance sectors use some of the most intense IT services and back ends in existence. The handling systems that process millions if not trillions of transactions per second. Not sure if the lack of technology is to blame for some of the financial black-holes that have been created through greed, avarice, poor management and other issues. Regardless of technology - markets at points can be abused, chaotic and driven by individuals.

As for a web 2.0 of finance - its already here - individuals can move money instantly from complex derivates funds to treasury bonds to obscure penny stocks all at the click of a button before a morning run all through an interface like Etrade - you are right however in comparing first world banks with developing countries banks. Markets with high internal financial service competition usually spur innovative uses of technology which in turn provide better and lower cost services for clients. The problem in todays markets might have been technology itself that moved money too quickly, letting individual risky investments group too quickly through technology for human questioning.

I guess my point is... when you make derivatives and and other financial products, you're taking a bunch crap and aggregating it. You hide this information and then send it off for someone else to buy and sell again to their clients. There are a few things happening here. There is "friction" in that each of these big investment banks and commercial banks are taking fees... also, by the time it ends up with you, you really don't know much about what you're buying. I agree that the technology of sloshing things around is pretty good and there are a lot of transactions, but there isn't a lot of technology to provide transparency and rigor and there are still tons of hidden fees for services you really don't need.

I realize you can do a bunch of stuff on etrade, but you're paying fees to fund managers, fees for bank transfers, etc.

FX is a good example too. Right now you're paying a very high spread rate if you're a small player, but there really isn't a good reason you should have to. You can easily check the real exchange rate and you should be able to just find a person/organization to swap with without going through some bank.

I was thinking about some of this the other day when someone sent me some cash via paypal but in Euros. I've also received payments to paypal in GBP and now when this happens I'm prompted asking if I want to convert it to USD or keep it in it's current state (Euros, GBP, whatever else I guess). I opted to keep it in the original currency because I don't have a lot of (shortterm at least) faith in the USD holding it's value. It didn't look like there was any kind of charge for the conversion, though admittedly paypal gets confusing a lot so it's possible there's tons of charges involved, but it got me thinking about a system like this but much better.

Like you I've gone to banks and tried to convert one form of cash to another and it's a huge pain in the ass. They make up their own conversion rates based or rates from hours or days prior, tack on extra fees, and often you have to wait or call in advance. Unfortunately with paypal I couldn't say take X amount of USD and change it to GBP, but if I could that would be huge.

It seems like a fairly simple idea actually, especially as more and more of this gets opened to the public. I'd gladly pay a monthly fee to a company/bank who I could send deposits to in one form of currency, and then online or at an ATM decide to change it to something else on the fly. Wonder what the legalities of something like that are.

As a scientist your comments regarding the subscriptions to peer-reviewed journals did ring a bell, and I can not but agree with your thoughts. A large part of the problem has been the unrealistic grading by rating agencies, and I just wonder if a closer monitoring of their activities by the government should have been enough to avoid this situation.

As you just mentioned about shopping around banks for the best exchange rates, I can recommend you XE Trade, which I have been using for a year and have saved me quite a bit while repaying my student loan: this at least has been my experience, and goes on the direction of taking advantage of the information available in the internet (compared with just walking around the city's financial district) and benefiting from the economies of scale and cot savings of a web-based lean organization.

And no, I do not work for them. I work in a research center in logistics and by the way we met personally about a year ago when I worked as COO in Mobuzz TV, and we have a swimming session pending together, whenever you are around.

Oanda is way ahead of the curve on this one. Easy to use very low cost international transfers and minimal spread exchange transactions as well as full trading capabilities.

The next step in this game is analytics for the masses. As soon as markets go all webly the webbed will need to exchange models of markets and performance. Being ready for that is a significant opportunity.

A few quick thoughts:

- The reason why you pay more as a small player on the forex market is volume. Matching millions of little orders so that buyers and sellers meet with exact amounts requires higher transaction costs. Small brokers sometimes have better rates but the risk of fraud (i.e. the trader uses your order for his liking) is higher - and if your order goes to straight-through processing then the broker as good as any bigger player.

- Why do banks even exist? Well, this is the first question in Finance 101 - the short answer is that there are several advantages to pooling - not least in transaction costs - you save an enormous amount of time, have much lower risk and, yes, banks do actually have some very useful expertise in execution. There have been many trials to conduct banking services in open markets but most of them were complete failures. This also goes for investment banking services (some companies auctioning off their own shares).

- When people hear derivatives now they all think of hugely complex stuff no one understands etc. The truth is that the current crises has little to do with derivatives (which here were just the means of making things transactable) but rather with a mismeasurement of risk (which to this day isn't solved and much more complex than most people believe) and a basic failure to see the systemic danger of pooling a certain asset class (very much exasperated by accounting standards).

I think the most important thing for people to understand is that banks deal with risk - that's their primary function. Without them pretty much nothing in an economy works. The history in this regard is very interesting - pick up any good book and have a go at it.

I agree, there is a lot of work to do on all levels of the banking system. Even the basic job of moving money around takes way too long and is still error-prone. The European Commission and Paypal seem to be the only players interested in actually doing anything about this fundamental problem.

This lack of a decent way to move money around quickly is what stops small-scale FX transactions from being as efficient as they should be.

As for trading risk, which is basically what managing investments and lending are about, it looks like the banks just don't have the right model for it.

That in itself is forgiveable - every organization and sector makes mistakes - but the problem is that a bunch of people made out like bandits whilst the whole thing was caving in. (When other sectors cave in, for example the music industry, the big bonuses generally stop flowing.)

Think about it. If bankers were so convinced that property lending was so low-risk, why were they being paid such big bonuses for managing the risk?

I think it should be possible to slim down costs in the money business, at all levels. But I wonder if the current round of consolidation will really do the job. I think lending needs to be more multinational and sector-oriented in the future, rather than being based around countries.

Crowds like Santander are just wedging banks together rather than changing them in a way that really improves the risk profile. Commerce has changed, and banking structures need to change too.

Interesting comment regarding currency exchange; there was a local startup here in Vancouver at a recent "Launch Party" event that was trying to set up Web 2.0 peer-to-peer currency exchange --

I chatted with them a bit and it seems like they have a solid business model, but they're not quite up and running yet. Might be one to watch though...

Very interesting discussion, all though I have never had money enough to really care about this :-) However I would love to see some move disruptive services in the financial business. Do you know this - a p2p credit and loan service.

This is an interesting topic. As someone who works in finance but doesn't understand it fully, I've also thought about what the age of the Internet could do to the current financial market landscape. Do we need huge institutions or would a large number of smaller players do just as fine?
Some ideas.
There's something to be said for having a "reputation" in the market place, even though this is a destructive notion at times. Financial institutions sometimes buy or sell instruments that require a continuous stream of payments to be paid between two partners, for instance interest rate swaps where one interest rate is swapped for one another. For their lifetime (until they expire), such instruments mandate continuous payments between the counterparties. "Trust" and "reputation" makes these transactions a lot easier since you may know that your counterparty is a huge bank with billions of dollars in the bank, so it would be somewhat hard for your one transaction to bankrupt it, but if you buy these from an anonymous individual on the internet, how do you know the risk associated with buying from him? He might go bankrupt instead of paying up on his commitment to you (this is termed "counterparty risk"). So you need a way of confirming the solvency of the individuals or tiny institutions you trade with, in the place of big reputations and public balance sheets.
Also, most nontrivial financial market players are subject to huge amounts of regulation, and have big compliance departments to ensure that all the ever-changing rules are followed.
These are all obstacles to shrinking the financial institutions below a certain size but maybe they can be overcome? It would be exciting, to say the least, with a world where individuals could issue instruments and trade just about anything from their living rooms. I think the technology is very much here, we just have to make it work from trust- and regulatory standpoints (and maybe others).

hello Mr. ito

thank u for the share of ur experience

i wanna konow how a transaction friction happens

would u mind replying with a sample case for my futher understanding?

looking forward to hearing from u..

jason with best regards


Light bulb indeed! Very interesting... what are the essential services provided by "banks" (investment houses, etc.), and how much of this can be replaced with online processing, marketplaces or clearinghouses? Also, what are the significant market inefficiencies introduced by banks?

For instance:

Distribution of risks & securities Skewed judgment of values

Perhaps an online marketplace/clearinghouse with independent/trusted risk-return assessment could potentially be at least as efficient in distribution, and remove the valuation bias inherent in a sales model of distribution.


So I agree that there are many important functions that banks serve. I guess I just think that unbundling them and making things more transparent is a good thing.

I do believe we need analysts, just as we need journalists and peer reviewers for academic journals. In the past, since distribution/transaction costs were so high and the job of distribution so difficult, the distribution channel typically paid for these people. However, the printing presses and telex machines for wire transfers are being replaced by nearly free IP packets. One question is how society should provide for these experts... I'm don't think amateurs can completely replace these professionals, although they can contribute and definitely take over a huge percentage of what is currently done by professionals.

I think that selling information and analysis as a service is one way. I think that institutions that focus on providing services for a direct and transparent fee instead of sticking into fund management fees, transaction fees and other hidden costs is probably one good way to go.

I think that bank fees have come down, and accessibility is much better than before. My bank, a very traditional (as in "normal") bank, offers me full access via a web interface. I can buy and sell shares, currencies, whatever, as I please (within my financial limits of course), and the price for that, well, it's still too much for someone who does not really have much money left for investment, but it's still a great improvement over what we had ten years ago.

It is probably easy to underestimate the infrastructure needed to really do financial transactions reliably and securely - it's not a trivial matter; errors could have far reaching consequences and even, at the extreme, destroy entire economies. The incentive for fraud is high, as well. So a certain inertia and erring on the side of caution is probably not entirely a bad thing.

Joi. Couldn't agree more. Indeed since the late 90s I have been focused on understanding just exactly how ubiquitous information and exponentially lower transaction costs would transform and shape the financial services ecosystem. Most of that time was working within the 'system' - as a senior executive in capital markets at an investment bank; but the persistently slow pace of adoption and the painfully protectionist/reactionary mindset of much of the industry (esp. wrt business model innovation) catalyzed my departure from the mainstream industry as I thought it would be worth trying to affect change from the outside having seen and experienced first hand the difficulties of trying from within. In 2005 I even penned a short futurist scenario (later adapted to a short video which can be found here: to spark a conversation in the industry but whereas it was viewed many many times and I had many compliments on my 'out-of-the-box' thinking, all to often industry insiders dismissed the main ideas as fantasy and not worthy of serious consideration. Indeed frustration at this response was a key driver in my decision to try a different approach.

One of the main obstacles I found at the time was the fact that the 'business' was making boatloads of money - at least on paper - in fact, things had never been so good. So when I would suggest that our industry's business model was bankrupt, well it's not so hard to understand why a polite pat on the head was about all the acknowledgement my ideas were going to receive from my peers in the executive committee.

I didn't predict the financial implosion of 2008. However anyone who has worked with me or read my (pre-blog) published articles or my blog will understand when I say I wasn't surprised. And although I certainly don't take any joy in the real economic pain it has caused - esp. to so many who have been collaterally damaged despite having done nothing wrong - there is (potentially) a silver lining.

The hubris I described above is gone. The psychological barriers to innovation and change in the industry are severly weakened. The opportunity for transformative disruptive business model innovation in financial services is now both real and possible. Time for the (past) leaders in financial services to dust off their copies of Schumpeter and make room for the new.

I don't know exactly what the next decade will bring, but I have a few ideas - many similar to the ones you expressed above - and I'm sure it will be a fascinating time.