The Slow Introduction of Open Banking and APIs in Japan

Welcome to Pacific Exchanges, a podcast from the Federal Reserve Bank of San Francisco. I’m Sean Creehan.

And I’m Paul Tierno. We’re analysts in the Country Analysis Unit. We monitor financial and economic developments in Asia. Today, we continue our ongoing series on fintech and digital innovation in Asia. We sat down Toshio Taki, the co-founder of Money Forward, a fintech firm involved in personal financial management, and an advisor to Japan’s Financial Services Agency on the ongoing efforts to promote a more open banking system in Japan.

Toshio gives us a sense for the ways Japan is breaking global ground really in promoting open APIs—also known as application programming interfaces—in its financial system. And he gets into some of the distinct drivers of Japan’s fintech scene and the particular challenge of providing digital solutions to customers that are really reliant on old-fashioned cash.

Yeah, he sees an open system for sharing data as enabling a more innovative future for Japan’s financial system, where the defining feature of a bank is no longer how convenient its ATMs are. Ok, let’s hear from Toshio!

Well Toshio, thank you so much for joining us today.

My pleasure, thank you so much.

You’re the co-founder of Money Forward, a fintech firm involved in personal financial management. Before we dive into the general topic of open banking and APIs, could you tell us a little bit about what it’s like operating in Japan’s fintech space? And knowing that you previously spent time in the US and are familiar with Silicon Valley, could you talk a little bit about the difference between Japanese and US fintech sectors?

Money Forward is a company that was founded roughly seven years ago. We started off with the service same as our company named Money Forward, which is a personal finance app, which was at the time heavily inspired by Mint.com in the US. There were some account aggregation services focused on personal finance since around 2001 in Japan, but at the point of 2012 there was not one platform that allowed users to use fully on the smartphone.

We initially didn’t start off branding ourselves like a fintech company. Probably at the time of 2012 the word wasn’t evident probably anywhere, but since we became a strong protagonist of open banking and banking API, since around 2015 we started to brand ourselves as fintech company representative of Japan. Because in Japan, when you say fintech it is not about, like peer-to-peer lending or more of like the payment app. We kind of focus much more on how the data is used maybe because there’s a lesser pain inside the economy of gaining credit from the banks, or lesser pain, it might be invisible pain. But since a lot of the population is happy with the cash based economy, we focus much more on the evident pain that people do not know their entire view about their money.

So, I think that kind of created a harmonious mood between banks and startups, because banks also needed to have those kind of data related solutions and we became like a very affordable option to deliver the same kind of service which would be much more convenient for the ultimate depositor.

Toshio, you touched on something in your last response about Japan’s preference or reliance on cash and I was wondering if you could talk a little bit more about how that affects a firm like yours and the broader development of fintech, considering that fintech is the shift away from cash towards more digital payments. How does the country’s affinity and reliance on cash affect it?

So there’s this very stereotyped understanding that Japanese people really, really love cash, and I don’t really necessarily say we love the cash but it is at least pretty much preferred because – the textbook answer would be Japan has been in long time, mainly non-deflation but at least in a very stable price levels. ATMs in Japan work insanely well, so most of the ATMs when you go to the store are kind of free, and all of the machines are moving and it actually gives you a mint paper condition. It is actually located everywhere in the city.

It depends on which measurements you do, but we always say 80% of the personal consumption is done in cash based payment. I don’t think there is similar developed country that operates at this level. So that is like the textbook situation of Japan. I think there’s another issue that there’s not one dominant known cash payment brand that exists in Japan. So Japan is kind of famous for having electronic money card, one representative is Suica, which is electronic money that could be used on most of the public transport.

But Suica is owned by everyone but does not facilitate peer-to-peer payment function. And also like if you try to go out for lunch, the odds that you can use Suica is around 20 or 30%, so you always need to carry cash. That is kind of totally different from what is going on in China where you have Alipay and WeChat Pay, at least everyone has either of their accounts and most of the merchants have been acquired by both of the networks.

I think the lack of this dominant network problem is a very, very large scope, especially when you think as a consumer which kind of electronic money you want to hold out when you go out. I think that’s a pretty large problem that Japan still faces. For us I think enhancing not just mobile payments but cashless payments on the whole is a very, very good thing because payments done by cash, if you want to manage those data, then you have to manually input those inside our app. I think it takes a lot of effort to do so and so we want to automate everything, not just personal finance but also corporate accounting and other things.

So we are a strong supporter and promoter of cashless payments and that’s how we have been trying to develop many policy recommendation in the past.

Just quickly a clarification question, when you say electronic money, is that referring to new efforts to create virtual currencies so like a virtual coin that is accepted by all major banks but held maybe by a different custodian? What exactly do you mean by that?

Electronic money I refer is very much like Alipay or Wechat Pay where you do have a virtual currency but it’s not like necessarily those crypto assets related things. So, I think a very normal shape of electronic account or wallets that people have, who could charge their money from their bank account or sometimes reimburse from them. Those kind of, some kind of a stored value facility is what I have in mind.

In addition to your role at Money Forward, you serve as a director of the Japan Association for Financial APIs, and just promoting the use of open APIs in general. Could you tell us a little bit about that push for a more open financial system in Japan? We’ve looked at this issue in other parts of Asia and so we are just interested in general in regional standards that might be emerging and where Japan is.

Since June the first of last year, of 2018, the Japanese banking act was amended to generally promote open banking. What I mean by that is, they introduced a very similar regulation to PSD2 (Payment Services Directive, revised) in the Euro Zone. Although there’s a large difference between PSD2 is that Japan does still not yet have a clear data portability related regulation or law. So a lot of the components of open banking is much more of voluntary effort to make the banking service much better.

But, since the Japanese government held a KPI (key performance indicator) to have most of the banks open up API by 2020, there are like 140 chartered banks in Japan and around 130 have a schedule to open up APIs by mid-2020, so around a year-and-a-half from now. Also, companies like us in the jargon of PSD2, we are positioned like account information service provider or a payments initiation service provider. So these kind of middle man companies are obligated to register under the Japanese Finance Services Agency (FSA), and we are regulated somewhere between a normal IT company, and a bank or electronic money operator. That kind of creates a room where you can also preserve the room for technological innovation, but at the same time, not everyone can operate those services so you have to hit certain hurdles of cyber security and information security related self-regulation.

And lastly, the Japan Association for Financial API is a self-regulatory body which is consisted of many of these user company of banking APIs. So, the members are all registered to the Japanese FSA. There are two very large goals. One is to create a solid self-regulation because with some fear towards cyber security, none of the innovation will happen anyway. And the other goal is to have a continuous improvement in the banking API arena, because although there’s at least one clear cut model of what is optimal as a banking API now, maybe those standards will be obsolete in two or three years where maybe more biometric authentication might come in place or maybe there will be other use cases of read-only data.

So we’re trying to have a dynamic regulatory effort. It is supported by both government and our local banking industry, so which is an evolving organization in short.

Just a quick follow-up there, how do you envision- if you have 130 of the 140 largest banks opening up their APIs, what kind of services do you see being supported initially? I mean you talked a little bit about maybe cash is too convenient so there isn’t as much demand for digital payments or peer-to-peer lending because of a healthy access to credit. So what do you think is the low hanging fruit in terms of services and products that may come about because of the open APIs, I guess your product would be one of them?

So there’s around like 20 companies already registered under the banking act and I think the initial use case is account information service providers. Our company actually operates multiple levels of that. So we both operate a personal finance app that uses read-only function of the retail banking account. And also we operate like a cloud based accounting software that also makes corporate accounting. Actually also look at us, but all of our rivals have started to register in the same regard.

So initially personal finance and corporate accounting seems to be the very basic usage of banking API. Now, moving on to payments arena, there are a few companies that are trying to utilize API to create a different interbank peer-to-peer payment platform. Japan has operated one kind of interbank operating, interbank payment system for the last around 45 years, but actually, there’s a lot of technological amendments that probably need to come in place and some of the private parties have built up a consortium built up of many banks in Japan, and trying to build an alternative network especially focused towards retail or individual peer-to-peer payment. Those will be coming out pretty shortly.

There will be some other players who have a focus to charge money to their own electronic money account, and recently there has been a very strong push inside Japan to promote electronic money this year and the next. So I believe those will be another player who will be trying to heavily utilize the banking API.

That’s the state right now, we do not have a very clear vision of what will be happening in two or three years from now.

On the topic of APIs and the 130 banks or so that in the next year and a half will be opening up APIs, can you talk what you do now in the absence of open APIs, to get that data?

So, as like in many parts of the world, we basically web scrape a lot of the banking accounts. So it’s like area where Yodlee has been operating since 1999. So basically users will entrust us with their read-only IDs and passwords inside our server. We believe it is a necessary procedure because we need to aggregate the data on their behalf but we always say, it is far from optimal because it will create a kind of a pretty dangerous situation in means of cyber security.

At the end of the day, there has been many players who screen scrape banking accounts in Japan but there hasn’t been one issue that they have become a victim of those cyber-attacks. But we feel the need to quickly move this situation into a much more token-based, authentication-based API economy.

Currently we are on the stepping stone, and although the users actually are pretty much happy with web scraping services, we believe in the case of any ultimate incident, we have to move quickly into much more secure economy. So things are kind of rolling up to be okay, but in one year or two years’ time from now, things will be much more different in our security procedures.

Sure. And with any kind of new technology or new way of doing things, some of the early players generally get to write the standards and guidelines for how those processes work. I think that there might be a sense that Europe is leading in writing open banking standards. Would you agree with that and that Asia needs to play catch up, or do you have a different view of how open banking standards are being written?

There’s multiple issues, we don’t necessarily think one country or one region is ahead in means of standardization right now. There is another effort, a global effort to create this movement, mainly at the open ID foundation, at the financial API committee. But looking at the discussion there, I think it is very hard to actually standardize a banking practice which actually is very much different between many countries. If you see on the local level of API, many of the API-related efforts happen on the local economy rather than as international remittance. It is actually very hard to have one standardized format for what a banking should look like.

Probably different countries will come up with certain different levels of standardization inside a country. There’s another like a little painful issue that there are multiple standards inside Japan regarding API infrastructure for the banks is developed by probably three to five system vendors. So they all have loosely different standardized formats of what the data should be, but I actually do not have a strong concern on this because at the end of the day, if you want to standardize the data formats and then you can just write one open source based wrapper of a function where it will simply standardize the dialect of those data.

I think that’s like the situation where we are in. I think the more important part is even if the government strongly promotes open banking, at the end of the day, how many depositors are actually using those APIs is the ultimate KPI. And as I mentioned, probably something that is optimal might be very different in two or three years from now. Just taking one function of let’s say old based token system. There has been multiple amendments and actually discussion on what is the optimal authentication system. So if you think about those things, I believe this is a little early to talk about harmonize or standardize formats of what the API could be. I think much more fruit comes from early practices and the learnings of what’s good and what’s bad.

So Toshio, beyond just the potential for different standards on APIs in different countries, wondering more about other related issues. So the emergence of data protection regimes like GDPR (General Data Protection Regulation), or related issues around cyber security, data localization, we’re seeing a lot of different countries in Asia start rolling out regulations and laws around data localization. Just wondering, is that another live issue that could impact the potential emergence of cross-border players, fintech players that are in theory going to leverage open APIs, because even if you find a way to harmonize or somehow account for data differences, just in that area, do these other laws come into play?

I think there’s again multiple issues that the Japanese government currently is trying to build something between probably the US and China and the Euro Zone. I think there’s probably like, looking at the recent announcing from the Japanese government, we are trying to have a more free flowing data economy. At the same time as I mentioned, Japan still do not have a stated right of data portability, so we I think in the next two or three years will be stating what privacy could be defined and especially in the digital era of the many economy.

The Japanese privacy act was only enacted 15 years ago, so before that time the word privacy was there but it wasn’t stated as like a very visible human rights issue. But, probably we are starting to catch up with how we think about privacy in the terms of Euro Zone, like the issues like the right to be forgotten or many AI related profiling issues and so on.

I think there will be some serious debates related to the privacy issues coming in one or two years from now. But actually we do not have a very straight forward view about how the international flow of this data will impact our lives in the Japanese banking API concept because as I mentioned, a lot of the data is very much local, and of course if there are multiple players who can gather data from outside of Japan and could provide a much more convenient service, that would be a very evident scenario. But, I believe not a lot of effort towards financial data has been done yet in Japan. So I think it is still in the early stages of what are the risk could be and what needs to be taken as a regulatory procedure.

So then looking into the future, what do you see as the impact of more open banking on traditional institutions and new startups? Does a traditional institution like a bank, who maybe has best in class service, do they have a strategic advantage versus a start-up that might be a bit more nimble and smaller?

Looking at the banks, it might be similar in many parts of the world, but a lot of the regional banks have a very similar level of operating standards. But when you look at some of the polls or questionnaires of why did you choose this one bank to bank with, 80% of the Japanese population answer that the ATMs are near, because they know there are like three or five times the demand. And that probably in the world where we don’t have cash anymore, those wouldn’t become the differentiators in like 20 years’ time from now. Probably how you could create information through a local connection or how you could create better advice based on human touch will probably become the differentiator. Not all of the banks are in the position to do so and that’s where we believe API will come into play.

So thinking that the banking industry is actually made up of two very large elements, one will be a secure infrastructure and the other will be the service layer, and probably from the consumers’ standpoint, people are paying less money towards secure infrastructure and would be embracing much more value from the service level.

Banking API is a very good way to enhance the service layer and what works out as the result of the product market base done by the startup company, and we believe that the players who are open to those approach will definitely have a very good role in 20 years’ time inside the banking industry.

Now, moving on to how startups could be different in future, I think there’s multiple roles but we always believe that many of the models in Silicon Valley or maybe in the US takes somewhat of a disruptor approach, where they create a very strong user experience and basically try to alternate some kind of a banking function. That hasn’t been place in Japan since there hasn’t been a strong credit crunch inside the SME (Small and medium sized enterprise) market, and also like a lot of Japanese citizens are kind of happy with the banking service. So I believe at least from the view of the economy right now, we believe startups that have a much more harmonious approach with the banks cascading and the recent many fund raising activities were done by very large banks or like telco companies and I think as of now those kind of partnership model is working out very well in Japan.

But again, things might be very different in three or five years from now, so I’m kind of looking forward to what will be happening there.

So maybe we can ask you to leave with a forecast of where you see open banking in Japan or more broadly in the region or globally going in the next five to 10 years, and what you see as the biggest barriers? At least in Japan it sounds like from a legal basis and just the perspective of getting the largest institutions to buy in, like that’s happened, and will have happened over the next year and a half. So what do you see as the future?

I think there’s like three very large elements. One will definitely be about how we involve the cashless economy into the differentiator factor of the banking industry. So, when people choose banks not because of the nearness of the ATMs, probably a lot of the focus will come into place on how the bank organic apps are performing and also what kind of exclusive products the banks have when you compare large banks or some very competitive internet based banks and I think there will be a strong competition towards what a really cool banking app could look like.

We sometimes benchmark J.P. Morgan Chase’s thing because we think a lot of the banks in the US are succeeding in creating their own organic app with a very good user experience and probably there will be some race towards there, which will also be competing with startup apps because well, in short startup apps are kind of just trying to be pretty cool. I think that’s one dimension of it there.

The other point is it is highly speculative, but maybe there will be some different kind of economic cycle in Japan where it will evoke some kind of different credit creation channel. What I mean by that is like, after the financial crises like in the US or in the UK, peer-to-peer lending came into place because there was some kind of shortage of credit at the time and probably that market hasn’t become evident in Japan but will be so because there will be some kind of a credit cycle within like 10 years. I think that will be another play.

The third is about whether banking will be different in five or 10 years from now. So as we have seen in China, maybe banking it will not be an optimal place for the retail financial services, maybe electronic money or some kind of financial platform might come into play. One move inside Japan is that probably the current Diet is trying to pass a rule that we can now receive our salary in electronic money instead of banking deposits. In Japan there is a regulation that you can only receive your salary with either cash or bank deposit, but probably there will be another exception of electronic money, which will create a room for the online platforms to become the first portal of financial decisions.

And again, Money Forward is another app that is trying to become the platform for peoples’ monetary concerns. So we’re trying to play very well creating some user experience in that regard, so that’s like the three visions I have.

Well this has been great Toshio, thanks for joining us today.

Thank you so much.

Thanks Toshio.

For more episodes like this, you can find us on iTunes, Google Play, and Stitcher. If you like what you hear, please leave a review. Feedback from listeners like you will help more people find us. And for even more content, look up our Pacific Exchange blog available at frbsf.org. Thanks for joining us.