Have Facebook Messenger Chatbots failed for Finance? +16 key Fintech developments

Hi Fintech futurists --

In the long take this week, I try out a contrarian point of view on personal finance chatbots. Trim, a savings chatbot, just withdrew support from Facebook Messenger. While lots of other chatbots are still invested in conversational banking, what could we take away from the counterfactual of chatbots failing to get B2C traction? What is the impact on the rest of the platform wars waged by Amazon, Google, and Tesla for connected homes, cars, and the Internet of Things?

The latest key updates on Fintech bundles, Crypto and Blockchain, Artificial Intelligence, and Augmented and Virtual Reality are below. Thanks for reading and let me know your thoughts by email or in the comments! Last but not least, these opinions are personal (or maybe made by a robot) and do not reflect any views of ConsenSys or other parties.


Long Take

It's important to know when you are wrong. And when it comes to things like technology platform shifts, what we are really talking about are Black Swan events. Such events are by definition not within the set of likelihoods that come out of a normal distribution. They are not the type of pattern our monkey brains are used to extracting from daily life. They are abrupt, alien, and surprising. They are the unknown unknowns.

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

For example, who knew that training a machine learning algorithm on images of cats and dogs for a decade would lead to Deep Dream, the second picture in the row above? Once you teach a machine how to see, you can teach it how to hallucinate the things it sees into everything else. And once you teach a machine how to hallucinate, it is not that hard to hallucinate fake people by the millions, as in the last picture. And once you can have fake people, tech companies inevitably do things like this: Snapchat and TikTok Embrace ‘Deepfake’ Video Technology Even As Facebook Shuns It and Snap reportedly acquires a deepfake startup. Are the people on Tinder real?

Anyway, one of the things I try to do is apply analogies from prior platform shifts to current ones. But it is easy to miss, or get irrationally optimistic. The only antidote to wrong thinking is measurement and data, especially data that contradicts your point of view. Which is why I was startled to see this in my Facebook Messenger last week:

No alt text provided for this imageNo alt text provided for this image

Trim is a chatbot software focused on personal finance. In the evolution of the sector, the mental model is first to (1) aggregate data through PFM software like Mint.com or Plaid, (2) analyze the information to understand someone's financial sitation at the current time, (3) generate a recommendation to improve the financial situation, and (4) implement the recommendation as an advisor. Trim implements savings improvements and subscription cancellations based on your transaction history, and it started out doing so through third party messaging services -- in particular Facebook Messenger.

Now lots of things grew up on Facebook, only to move away into independent companies. This is the story of the social investing / trade following network called Kaching, now called Wealthfront. But I had thought Messenger to be different -- not just a starting channel, but the actual paradigm. The core argument is that the chatbot and voice channels are platform shifts, they are the place where customers are going to be in the next decade as interaction preferences change. Devices like Amazon Alexa and Google Home will be the aggregators of user attention, and power conversational banking. Therefore, the banks themselves become subsumed to the platform into which they are integrated.

There still remains lots of data that proves out the decline in bank branches, human customer suppport, and an increase in Millennial and Gen Z preference for mobile apps and interaction.

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

There still remain multiple conversational banking companies that are trying to build out direct to consumer services, like Plum, Cleo and various other primarily-European Fintechs. And the Trim data point is just one of the few symptoms that the current generation of messaging-integrated chatbots are not having meaningful B2C impact. Perhaps they are not easier to use than mobile apps. Perhaps the conversations are too frustrating and the technology is too early. Perhaps Facebook has ruined its reputation, between the election cycle and the crypto-currency launch.

But we need to consider the negative data points seriously to pierce our reality distortion bubbles. If chat and voice are not sufficient architectures for platform shifts, what impact does that have on the large tech companies which are battling each other over AR/VR hardware and smart car operating systems? In the Mixed Reality and IOT link sections below, I highlight a few industry bets that are dependent on consumer behavior changing to adopt conversational banking. Embedding Amazon Alexa into gas stations so that people can pay for fuel with their voice, or creating e-Commerce dashboards in smart cars, may end up being quite a waste of time if our basic assumptions about platform shifts are wrong.

No alt text provided for this image

I am not arguing that we are wrong already -- but that the probability-weighted chance of being right has decreased at least a smidgeon. Stay vigilant!

And let me know if you think conversational banking is here to stay, or just innovation theater.


Key Fintech Developments

No alt text provided for this image
  1. Citi’s investment bank plans to hire 2,500 coders this year

  2. The Canadian robo-advisor Wealthsimple has registered three new businesses across several provinces, including Wealthsimple Cash, Wealthsimple Payments Inc., and Wealthsimple Digital Assets Inc.

  3. Banks Elevate SMB Lending With FinTech Tie-Ups and Why 64 Pct Of Merchant Services Providers Want A Payments Overhaul

  4. Monetary Authority of Singapore Has 21 Aspiring Fintechs Apply for 5 Digital Bank Licenses -- and in particular, Mapping the Singapore digital banking contenders by their experience (sent to me and written by one of the reader's of last weeks' edition)

No alt text provided for this image
  1. Court order slows down SEC bid to obtain Telegram banking records tied to token sale and Ripple Labs’ Real Business Model: Selling XRP to Retail Investors

  2. Head of Russian central bank lets companies test stablecoins

  3. Bitcoin.com wants to raise $50 million through an exchange token sale and Europe’s first bitcoin bank has expanded its service to offer ether trading directly from users' current accounts. (some good Ethereum features over here)

  4. Rabobank has collaborated with IBM on the development of an application allowing consumers to support the farmers who grow the beans that make their cups of coffee.

No alt text provided for this image
  1. Report by Chinese Academy of Social Sciences said trobots and other AI applications will have taken roughly 8 million to 10 million manufacturing jobs from migrant workers, averaging 1.6 million to 2 million per year

  2. How AI Is Really Going To Change Real Estate In 2020 And Beyond

  3. TypingDNA has raised a $7 million Series A round; founded in 2016, it developed “proprietary artificial intelligence algorithms” to authenticate users based on how they type

  4. Worldline is to stage the first live demonstrations of a new pay-by-face system at the National Retail Federation, saying that it uses "3D facial biometry and respects the users’ data privacy

No alt text provided for this image
  1. PWC report on How VR and AR will transform business and the economy globally and in the UAE and 5 Million PlayStation VR Units Sold, Sony Announces

  2. French wearable company PKvitality raises €2.25M

  3. Elon Musk heads to Shanghai for first deliveries of made-in-China Teslas to public

  4. Alexa Payments Coming For ExxonMobil Customers and Visa: Road-Tripping Into Connected Commerce’s Future


Featured Interviews, Podcasts, and Conferences


Looking for more?

BBVA to sell chocolates on Amazon, while gamer hardware co Razer tries banking in Singapore; plus 12 key Fintech developments

Hi Fintech futurists --

In the long take this week, I continue on the thread from last week to show how weird and interesting the banking industry has become in 2020. We have BBVA partnering with Amazon to become a vendor of chocolate. We have a competition in Singapore for 5 new banking licenses, with applicants including gaming hardware company Razer, ride-hailing superapp Grab, telecom Singtel, among insurance companies and venture funds. Is it all nonsense? Or is there logic to the chaos?

The latest key updates on Fintech bundles, Crypto and Blockchain, Artificial Intelligence, and Augmented and Virtual Reality are below. Thanks for reading and let me know your thoughts by email or in the comments! Last but not least, these opinions are personal (or maybe made by a robot) and do not reflect any views of ConsenSys or other parties.


Long Take

Anyone watching Fintech over the last decade has recognized an increasing shift of power from product manufacturers to the platforms where those products are sold. In the case of Amazon, Google, and Facebook -- finance is just a feature among thousands of others. I've made this point since 2017, when Amazon launched lending into its platform. Brett King has been a bit more generous in the categorization, calling the shift "embedded banking". This means that banking products are built into you life's journey, not accessed in a separate customer center location. The financial API trend is a tangible symptom of this vector.

No alt text provided for this imageNo alt text provided for this image

So let's start with the assumption that this is roughly correct. Risk and capital live in various touchpoints across other apps, integrated through APIs. The other part of the framework to remember is -- who is able to deliver infrastructure and customers? For a future-of-payments report earlier in the year, Autonomous Research looked at the following key players as determinants into the equilibrium of payments services in a geography: (1) public and national banking, (2) private financial networks and markets, (3) telecommunications companies and hardware provicers, and (4) web/mobile apps and social networks.

No alt text provided for this image

These four power centers are each able to deliver financial products at the level of the individual, but depending on risk aversion, industry maturity, and cultural preferences have varying levels of power and influence across geographies. In Kenya, the telecoms had a unique advantage in building out a mobile currency across an underserved market. In China, mobile apps took on the challenge of creating a new method of commerce and banking as State-run enterprises focused on large institutions. In the US, national and private banking entities have enforced a strong oligopoly across services, which is slowly being melted by the tech firms.

So let's say you are BBVA, and you were *way* early to spend over $100 million on a Fintech bank acquisition, now your offering looks like this.

No alt text provided for this image

Then this is what your digital channels look like -- New Year, New You? Spain’s BBVA looking to sell financial services on Amazon. As I understand the news, BBVA will become a vendor on Amazon to first sell non-financial goods and services in order to get acquainted with the channel. Those goods are: a Michelin three-star restaurant’s recycling project, and Rocambolesc, chocolates and sweets crafted by a high-end confectioner. It's weird, super weird! But I like it. Without learning how to do the thing, all you have are strategy consulting PowerPoints.

No alt text provided for this image

In a comparable vein, the Monetary Authority of Singapore is planning to grant 5 new banking licenses to new companies in 2020. Of those, 2 are full bank licences that permit retail banking, and 3 are meant for wholesale banking. This has kicked off a bunch of public company and private equity-back bids by various consortia to stand up new banking power-houses. Here are a couple of the ones that crossed my RSS reader:

  1. Ant Financial applies for Singapore digital banking licence. This multi-billion dollar Chinese fintech company, built on the success of Alibaba as an E-Commerce platform and social network, is trying to become a wholesale bank.

  2. Temasek-backed firm leads consortium to vie for digital wholesale bank licence. A public financial company (Sheng Ye Capital), a capital markets firm (PhillipCapital), and an artificial intelligence software player (Advance.AI) are trying to build a data-driven "next generation" SME bank. Also, Temasek is the investment fund for the government of Singapore.

  3. Razer Fintech leads consortium in digibank bid. Ok, this one is led by a hardware gaming company (Razer), targeting the youth market. It also includes an insurance company (FWD), an Internet company (LinkSure Global), venture fund Insignia Ventures Partners, a vehicle wholesale marketplace Carro, and the founders of a large supermarket chain (Sheng Siong).

  4. Singtel, Grab Team Up To Launch Digital Bank In Singapore. On of Asia's largest telecoms (Singtel) and South East Asia's ride-hailing-super-app (Grab) are joining as the two bidders as well.

Let's visualize these four entrants on our strategic map from before.

No alt text provided for this image

This is the current state of play -- just for a banking license in Singapore! Every single consortia is a mix of multiple types of companies and customer segments. Further, in several cases, there are no traditional Wells Fargo-like financial services providers at all. What most of them do have, however, is a large customer footprint and a brand that clients love. Razer is well known and "cool" in the gamer market. Grab is the local version of Uber. It's the equivalent of something like Orange becoming a bank. Brands will increasingly grow financial product polyps on their audiences.

To approximate reality, we have to add dimensions to this example. For example, expand the licensing question from Singapore across geographies -- Europe, UK, US, Latin America, Africa, China. Or, expand the product segments from banking to investing and wealth management, to more complex lending like home or auto financing, or to various versions of insurance. This single starting point, however, highlights just how dynamic the ecosystem has become, and how non-linear and surprising the outcomes will be. Did you ever expect 2020 to start with this company applying to be a bank?

No alt text provided for this image

Featured Interviews, Podcasts, and Conferences


Key Fintech Developments

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

Looking for more?

7,000 bank branches shut down and 425,000 jobs lost -- melting Banking into a glass half full; plus 12 key Fintech developments

Hi Fintech futurists --

In the long take this week, let's make a collective decision to see the glass as half-full. While physical banking (7,000 US branches gone during 2012-2017) and employment in the sector (425,000 jobs lost since 2013) has been contracting, digital commerce, banking, and investment management have been growing. Even DFA is finally giving in and lowering fees on their $600 billion institutional mutual fund family. Of course, Fintech has been a slow and gradual transformation, not a rapid disruption. We can make a choice to bemoan the loss of the past, or a choice to express an excitement for the future and participate in its making. Which side are you on?

The latest key updates on Fintech bundles, Crypto and Blockchain, Artificial Intelligence, and Augmented and Virtual Reality are below. Thanks for reading and let me know your thoughts by email or in the comments! Last but not least, these opinions are personal (or maybe made by a robot) and do not reflect any views of ConsenSys or other parties.


Long Take

It's December 30th! Happy holidays and upcoming New Year! Do you really need a long take about Fintech today? If yes, here are the most popular articles from 2019:

  1. Google and Amazon's mixed reality advertising could become digital lending and payments platforms

  2. Fintech Price War -- Goldman's $1.3B Marcus burn, Neobank £200MM loss, ETrade's $75MM trading fees

  3. Softbank's $500M human experiment; a breakdown of N26's latest $2.7B valuation; Plaid acquires aggregator Quovo for $200M

  4. Ok, Boomer -- a meme for the broken political economy

  5. Deutsche Bank to fire 18,000 people while Amazon upskills 100,000

It's interesting to see what bubbles up in our collective, narrative Id. Anxieties about big brands (Google, Amazon, Facebook, Goldman, Deutsche), destruction and rebirth in the shape of layoffs and venture fundraising, and intergenerational tug-of-war along the axes of money, power, and social media. Not much has changed in our psyche, whether looking at the journey of Odysseus or Luke Skywalker (I am standing by the classics, thank you!). The eyeballs deserving our pointy sticks have just gotten bigger.

No alt text provided for this imageNo alt text provided for this image

We want to see underdogs beat giants. We want our version of the good triumph over our version of evil. But perhaps now we realize that which side you are on is often random chance, an accident of where you were born and how you grew up. And this is a very powerful realization. It implies that we can choose a point of view, rather than be tribally married to one by emotion.

I was lucky to go quickly from Lehman Brothers into the 2008 financial crisis, and then directly into Fintech. This meant that the narrative transition from the prestige of financial services to the inevitability of consumer technology came cheaply to me. Not much to lose, and everything to gain! For many other people, similarly situated or more senior in their careers back then, it was costly to look at software and say -- this will take my craft from me. This will take the thing I like doing, the thing I am good at, the thing that is my personal identity.

Even today, when the data is louder and clearer every day, you will hear push back, hesitation, and skepticism. The neobanks are unprofitable! Roboadvice isn't as big as analysts said it would be! All the cows are sacred! I empathize with this story. I want the good guys to win, not some Skynet abomination. And yet, why not just flip the narrative? Why not realize that it is more fun being the growth industry than the melting ice cube? Your role in the game is not fixed -- it is a choice still.

No alt text provided for this image

I could tell this story any week, because the symptoms are a recurring stanza of data. Let's lay out a few symptoms at the end of 2019. According to the Mastercard spending pulse survey, this holiday season (November 1st to December 24th) overall spending in US retail grew 3.4%, with eCommerce growing 19% this year after an 18% increase last year. While still in the minority, it is clear that eCommerce is the growth area, while brick and mortar growth is losing share. Which business model would you invest in today? Which payments company supports the more valuable activity?

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

You can aggregate the eCommerce numbers across the key end-of-year American holidays, getting around $15-25 billion of sales. Just one Single's Day in China stacks up to $40 billion across eCommerce platforms. The direction of travel is not a mystery, it is just not evenly distributed.

Here is another symptom. America's bank branches are closing down across the country, hitting rural towns especially hard. In a country where separation between the States and the Federal government has long kept local banks on life support, the mobile web is driving a nail in the coffin of physical banking. Across all types of geography, nearly 7,000 branches have shut down -- a decrease of 7% between 2012 and 2017. Some parts of the country have seen the branch-to-population ratio fall by 50% from 5.5 to 2.3 (see report). But for most of us, the bank branch is already in hand.

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

Just look at this chart above. One of the core arguments against digital banking and money is that it is regressive. People on the lower rungs of society's income ladder are underbanked, subject to high fees and mistreatment from our financial institutions. Physical cash is their anonymous and universally accepted saviour, goes the argument. And yet, nearly every human owns a smartphone and over 40% of those in the US use mobile phones for banking (back in 2017 no less!). Digital currencies like Bitcoin thrive in places that lack a stable financial and legal infrastructure, and Telecom enabled monies like M-Pesa are prevalent in Africa -- where there is more access to phones than electricity.

No alt text provided for this image

So it is not a great suprise, as artificial intelligence pushes on the value of knowledge workers and technology companies centralize attention into honey pots, that global banks have cut 80,000 job this year (reports American Banker). The last six years add up to a bit over 400,000 eliminated positions, with banks like Morgan Stanley bleeding about 2% of talent per year. Some, like Deutsche Bank, make more loud cuts. In order for Fintech to absorb these positions, it needs roughly $80 billion of funding -- assuming $200,000 of cost per fully loaded position. Given that Fintech today raises about $50 billion per year, that number is not outlandish. The melting ice cube could be filling up the Fintech glass.

It's not just people -- it's pricing and business models. The $600 billion AuM money manager Dimensional Fund Advisors ("DFA") pioneered indexing and smart-beta many decades ago. DFA founder David Booth amassed enough fortune to put his name on the University of Chicago School of Business. After ignoring ETFs and roboadvisors for as long as they could, the firm is finally cratering to pressure -- Dimensional Fund Advisors makes unprecedented slash of fees across all its mutual funds and declines to rule out DFA ETFs muscling into the crowded market. While the first version of mobile-first money management may not have eaten DFA from without, it has now eaten it from within (you can find DFA funds within the Betterment offering for advisors). Same for Schwab, Merrill, Goldman Sachs, and any other incumbent proclaiming themselves to be a technology firm.

I started out with the premise that you can choose your side. It may seem like an avalanche of price cuts, firings, automation, and closure. That is, if you sit on the side of the high fees, manual paper-based process, and an unnecessary retail footprint. Or, it may seem like an emerging financial economy that caters closely to the needs of the human being with a tiny super computer in hand.

Here's to a 2020 where the glass is half full!

No alt text provided for this image

Key Fintech Developments

No alt text provided for this image
  1. UK-based Fintech Curve Introduces “Curve Send,” a New Way to Send Money Between Accounts Free of Cost

  2. India-Based Fintech ZestMoney Secures $15 Million During Series B Funding Round Led By Goldman Sachs

  3. More Pressure on Advisors: Schwab's Latest Retail Offering, Automated Decumulation

No alt text provided for this image
  1. Total DEX volume on Ethereum surpasses $2.3B in 2019, with IDEX leading the market

  2. China’s blockchain cross-border financing platform is already being used in 19 of China’s 23 provinces and ‘Japan’s Amazon’ Rakuten now allows users to convert loyalty points to bitcoin, ether and bitcoin cash

  3. Hedera Hashgraph, the company behind the blockchain-like Hedera network, is asking investors to wait longer for tokens they paid for, in order to stabilize their cratering price.

No alt text provided for this image
  1. Amazon Seeks Patent For Hand Recognition

  2. New Patent Reveals Tesla Aiming For Quicker Autonomous Driving Improvements

  3. Some of China’s most prominent AI start-ups were put on a US trade blacklist this year, as the industry faces greater scrutiny from the world’s largest economy

No alt text provided for this image
  1. Tesla owners in China can soon play mahjong and poker in cars

  2. HaptX Secures $12M Financing to Produce Next Generation of Haptic Gloves

  3. 10 IoT Companies To Watch In 2020 and Why 5G is Essential for AI, IoT, and Robotics


Featured Interviews, Podcasts, and Conferences


Looking for more?

Top 5 Decentralized Finance predictions for 2020 and beyond; plus 20 key Fintech developments

Hi Fintech futurists --

In the long take this week, I revisit decentralized finance, providing both an overview and 2019 update. The meat of the writing is the following long-range predictions for the space in the next decade -- (1) the role of Fintech champions like Revolut and Robinhood as it relates to DeFi, (2) increasing systemic correlation and self-reference in the space, which requires emerging metrics for risk and transparency, and (3) the potential for national services like Social Security and student lending to run on DeFi infrastucture, (4) the promise of pulling real assets into DeFi smart contracts and earning staking rewards, and (5) continued importance of trying to bridge into Bitcoin. Here's to an outlandish 2020!

The latest key updates on Fintech bundles, Crypto and Blockchain, Artificial Intelligence, and Augmented and Virtual Reality are below. Thanks for reading and let me know your thoughts by email or in the comments! Last but not least, these opinions are personal (or maybe made by a robot) and do not reflect any views of ConsenSys or other parties.


Long Take

Decentralized Finance Overview

Happy holidays, good people! As 2020 sneaks up on us, 'tis the season for predictions and prognostications. Which makes me realize how much I miss being able to get into a meaty, quantitative analysis in a well formatted PowerPoint document. I realize this because several blockchain industry analysts have put out absolutely fantastic work that should be your reference for the coming year. If you read anything but this newsletter, you've got to read:

  1. Blockchain Capital's 2019 Year in Review

  2. Messari Research Crypto Theses for 2020

  3. DappRadar 2019 dapp Industry Review

So let's narrow our entire lens to decentralized finance. There's a lot to say about the overlap of Fintech and blockchain, but let's find the very edge of the frontier and focus right there. I am talking about financial manufacturing engines that sit on large, distributed, open source, programmable infrastructure. I am talking about global cyborg money, flying around between regulatory frameworks, finding the weak spots, and building cathedrals of complexity before anyone can notice. I am talking about systemic risk, and the promise of what is most likely the future of financial services after Google, Facebook, Ant Financial, and Apple dismember the banks.

To catch you up, here are the basics. If you already know the basics (you anarchist devil!) scroll down for the predictions below.

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

Ethereum is the leading platform for decentralized financial software, full stop. EOS and TRON (two massively funded competitors) are the Windows phone of this segment. If given the benefit of the doubt, those platforms have a much better shot at the gaming and gambling sector than to cross the bridge to real financial infrastructure. But being a leader doesn't yet mean celebrating industry victory -- Ethereum's decentralized lending and trading are floating around 2,500 daily users. On the other hand, there are only 8,000 Registered Financial Advisors in the United States and even fewer hedge funds, private equity firms, and family offices. So 2,500 users isn't necessarily small, and it just happens those users are moving real money in terms of volume (see the vertical axis in the apps above). You can tell I am of two minds on the topic. Here's a good Messari quote to sum it up:

No alt text provided for this image

What are these application categories that make up the primitives of DeFi referenced above? Below I walk through the major financial protocols by daily unique wallets, a proxy for popularity. The first use case is decentralized exchanges -- some straight-forward peer-to-peer trading. API integration into other apps is starting to become the killer feature for adoption. For example, if you have a payments app and need to trade in and out of an asset, integrating Uniswap is a quick solution. Notably, the exchanges do differ by where they store data and if/how they form order books.

No alt text provided for this imageNo alt text provided for this imageNo alt text provided for this image

The lending product above is a bit more complicated. In the case of MakerDAO, you put some Ether into a black box as collateral, and get issued a loan pegged to the US dollar collateralized by that box. If you withdraw your Ether, you get charged an interest rate determined by an open governance community. Other things like Compound are essentially a securities lending marketplace -- you get paid some market-clearing interest rate to loan out assets like Bitcoin, which someone is borrowing for a trading or hedging purpose. Having interest rates native to the crypto economy is very helpful, but we are still shy of lending based on actual underwriting and off-chain risk.

Payments is also spreading to Ethereum, which you can see by the growth in USD-pegged stablecoins. The largest one, Tether (putting aside its dubious history for now), has recently shifted its technical infrastructure to Ethereum and has been responsbile for a meaningful amount of transactional value. According to Blockchain Capital below, nearly 40% of the stablecoin market is Ethereum-based Tether, accounting for 80% of transaction value overall. DAI, which is the native stablecoin of DeFi issued by MakerDAO, remains a small percentage. I would guess this is because DAI was originally only collateralized by Ether, while USDT was purportedly backed by ($ billions in) cash. Notably, DAI is now able to accept multiple types of collateral.

No alt text provided for this imageNo alt text provided for this image

Outside of the core financial products, we have multiple services and exotic asset classes developing. First up is automated asset allocation with trading and rebalancing -- a wealth management service from Set Protocol.

No alt text provided for this image

Next, feel free to buy insurance mediated by software from Nexus Mutual.

No alt text provided for this image

Or if you want to get real complicated, here are a few projects that have been building synthetic assets and derivatives (see UMA and Synthetix). Nothing like some good old financial engineering to lever up a small market!

No alt text provided for this imageNo alt text provided for this image

I'd be remiss not to mention the broader concept of "staking". This feature will be built into Ethereum over the next several years, but already exists in various other blockchain protocols. The main idea is that by taking some part of your money and putting it up as a "stake", you get a native interest rate and some amount of control over governance (i.e., something between equity voting and dividend yield). This financial commitment also performs the function of securing the historic accuracy of the data on the chain itself, like electricity-consuming mining does in Bitcoin. A number of staking blockchains have already launched, and also offer a junk-bond type of interest rate. See Staking Rewards and the chart below, and Coinbase for an example of a large Fintech providing the service.

No alt text provided for this image

Soon enough, you will be able to go from start to finish across your financial life using these various services. A year ago, they were disconnected and difficult to use, unless you like installing browser plugins and downloading strange Internet software. This is now much less true. Given that crypto native financial products have proliferated, user aggregator applications are starting to emerge. Whether the leader is ArgentInstadappZerion or someone else, the blockchain-enabled Internet is about to get a whole lot more financial. It's like watching Mint.com emerge in 2007.

No alt text provided for this imageNo alt text provided for this image

So what does the future to come look like?

Top 5 Predictions for DeFi in 2025

(1) Global Fintech champions like Robinhood, SoFi, Revolut, and Square will enter and win the crypto race as soon as the business model is established

No alt text provided for this image

There's no obvious way for crypto aggregators to make money. They could try to scale to millions of users, and then charge a subscription fee. But they are missing the revenue streams that traditional financial institutions use to subsidize free services. For example, you could get free financial advice from the Schwab roboadvisor, but you will pay the management fee pn the ETFs from which the portfolio is made. You will also cede some interest income to Schwab on the cash that you hold in their sweep account. The set of licenses and regulatory entities that Schwab holds allows it to bank, broker, and manage assets and charge these market fees.

In DeFi, the economics of financial manufacturing are extremely uncertain -- and because they are open sourced, I believe they will trend towards zero. Once you invent a software and launch it as an open source project, that automation is out in the world. It's a protocol, not a company. Thus the user aggregators can be quickly copied and co-opted by incumbents. Look for example at the early "staking" competition between Coinbase and Binance, leading to an immediate price war around these returns, eating away a potential revenue pool for new companies. Further, actually touching client funds immediately makes these apps into regulated custodians or money transmitters or financial advisors, which means a lot of compliance cost for little upside.

On the other hand, there are already a half dozen mobile apps -- all Fintech unicorns -- that have licenses and millions and millions of users. Today, Square, Robinhood, Revolut and a few others offer crypto trading. This is most likely enabled with an omnibus account at an institutional, centralized crypto exchange. If the alphabet soup of DeFi projects becomes compelling enough for the users of those mobile apps, they will aggressively feature-compete. I also think the consumer Fintechs will out-race even the financial incumbents experimenting with enterprise blockchain, such as Goldman Sachs, JP Morgan, and Santander. If you raised money from SoftBank, you have to take on large risk, while the banks will take another 5 years to touch real DeFi.

(2) Risk management and regulatory transparency is paramount in DeFi, as leverage and systemic correlation threatens the ecosystem

No alt text provided for this image

A lot of this space is self-referential and recursive. That's fine if it is anchored by some powerful set of fundamental activity. Imagine the cigarette economy of prison. Or less controversially, imagine going to the moon and establishing a colony that develops a new currency based on moon rocks. Creating synthetic derivatives based off the moon rocks is fine, as long as people are doing something to earn those moon rocks. If they are just creating markets for borrowing rocks, and then taking the interest from the lending, and using that as collateral for more borrowing -- well, we know what 2008 looks like. For a good take on this Russian Doll problem, check out the Defiant.

What if, for example, one of the core protocols like Maker (pick your favorite one) blows up. Is that likely? Crypto's 2016 attempt at venture investing (The DAO) blew up in a spectacular hack, leading to an existential crisis with a $500 million cost and an SEC warning. Crypto's 2017 financial attempt at investment banking (ICOs) led to a year and a half long bubble, followed by a 90% price drop across thousands of assets, erasing $500 billion of market capitalization as a result. Crypto's 2019 attempt at banking and lending will surely hit some wall, especially given the sofware integrations and user concentration across these protocols.

Overall, I find this destruction encouraging -- each time, the industry gets it a little more right. More tools are created. Scar tissue is formed. Lessons are learned. To that end, the only antidote to dumb mistakes is transparency and data. When looking at high-yield future tech investments, maybe think about risk and not just return! What form does that risk take when looking at decentralized software? Who is your counterparty? What does the system look like? We've been staring deeply at this question at ConsenSys -- see the DeFi Score initiative. Certainly, regulators will be asking the same question as the space grows more popular. Data and analytics services that deliver good answers will weather the storm.

(3) The software of Decentralized Finance will be replicated by central banks and governments and launched nationally

No alt text provided for this image

Either this is super obvious, or I am completely insane. Nearly every nation on the face of the Earth is trying to understand what it means to launch a Central Bank Digital Currency (CBDC). Two 2019 symptoms have caused this rush -- Facebook's Libra, and the Chinese digital yuan. Neither of those initiatives are a true cryptocurrency, in the sense that they are neither trustless nor permissionless. But both are inspired and motivated by Bitcoin. As a result, Sweden has hired Accenture to create the e-krona pilot, the European Central Bank is exploring a CBDC, and countless other meetings are taking place behind closed doors to formulate a response.

That said, when I think about government-backed financial infrastructure, I think about the Automated Clearing House in the US, the Faster Payments initiative in the UK, and the PSD2 regulation of open banking APIs throughout Europe. Similarly, national regulation often mandates student lending access through public-private entities, provides social security or pensions, and some form of medical insurance. That is to say -- governments are definitely in the business of running communal financial infrastructure across asset classes. And if Bitcoin gets transmuted into CBDCs, perhaps DeFi can take the place of public services.

Imagine Fannie Mae run as a smart contract -- operated, or maybe audited and controlled, by the US government. It can combine attributes of collateralization from Maker with the services of a third party underwriter. This may be an "oracle" underwriter external to the whole process, or an embedded machine learning algorithm trained to assess properties. Alternately, imagine a smart contract that distributes retirement income in a tax-advantaged way, similar to the Schwab roboadvisor for older generations. However, this software is built and audited by the Department of Labor. Such a version of the world presupposes mass crypto asset adoption. It also assumes that the philosophies of DeFi have overlap with national interest -- something I believe we will have to solve in order to help people most in need.

(4) Real world assets will be assimilated into the DeFi Borg cube through staking incentives, and would be a big win for the space

No alt text provided for this imageNo alt text provided for this image

Look, I know you are bored hearing about enterprise blockchain and Security Token Offerings. Neither has had the short-term impact that many people hope, and there is an impression that institutions continue to sit on the sidelines. While it is true that Santander issued a bond on the Ethereum mainnet, supply chain consortia like komgo and Vakt are moving trade at scale, Chinese tech companies are installing blockchain invoicing across large geographies -- these developments have not excited the popular imagination. Similarly, STO and IEO volumes failed to match the hype of the ICO cycle. Yet this is a must-do step in the journey to economic adoption.

I will call out two projects. The first is Figure, a home equity loan platform that records and manages its loans on a proprietary blockchain. Figure has raised over $100 million, is run by the founder of SoFi, and has quickly turned blockchain into a feature, not the main course. The second is Tinlake from Centrifuge. Now that Maker accepts more than just Ether as collateral, Centrifuge is working on bringing other financial products as collateral for USD-pegged stablecoin issuance. Those could be anything from invoices to houses -- just put them into a crypto box.

I can imagine a world where staking is the primary source of interest rates on the decentralized web. When your money is sitting idle, you commit it to generate cybersecurity for everyone's financial infrastructure, and get paid interest. This could be a larger source of interest income, in the sense that it can absorb more funds for productive activity, than (1) a governance body setting an interest rate for issuing debt, or (2) a market interest rate for borrowing assets for trading. To reach that interest, you may want to commit traditional assets, from your brokerage portfolio, to your home equity, to even just a promise for repayment. This bridge may be built by the DeFi entrepreneurs. Or it may just be built by financial incumbents as they spin up institutional chains, analogous to Figure. The downside of a global, default risk-free rate is that it is a tax on those who are unaware it exists. When everyone else stakes, and you do not, it erodes your purchasing power through inflation. Sound familiar?

(5) Don't bet against Bitcoin, and try to pull its value into DeFi machines.

No alt text provided for this image

That chart says it all -- Bitcoin is still the most systemically relevant crypto asset. The more DeFi can figure out how to pull Bitcoin into the Ethereum fold, the higher the water rises for everyone.

Enough said! Happy holidays and thanks for reading.


Key Fintech Developments

No alt text provided for this image
  1. Congress Passes Sweeping Overhaul of Retirement System, Bill encourages 401(k) plans to offer products with guaranteed income payment and Schwab set to launch 'shoe-that-dropped' subscription retirement income robot that acts like a virtual annuity and produces 'predictable' paycheck. (love it!)

  2. Hometap , which provides a smart, new loan alternative for tapping into home equity without taking on debt, has announced that it has secured $100 million in new financing.

  3. Fintech lenders tighten standards, become more like banks and San Francisco’s Credit Karma Acquires Fintech Haven Money, a Savings App.

  4. Lemonade and the Art of Spin -- a critical take on the economics of the Insurtech unicorn and Lemonade is now a GEICO Insurance Agency partner. The latter has referred approximately 5,000 visitors to Lemonade’s site in November.

  5. Citi Treasury Services, PayPal Team On Global Mass Payouts

No alt text provided for this image
  1. Exploring DeFi trading strategies: Arbitrage in DeFi (was new to me!) and Nic Carter's On-Chain Data Story and Are DeFi's Money Legos Creating Systemic Risk?

  2. Ripple Raises $200 Million as Part of Bid for XRP Adoption

  3. Cardano Now Has Over $195 Million in Staked Digital Currency (ADA) on its Test Network

  4. Fidelity Digital Assets intends to support Ethereum in 2020

  5. Swedish central bank taps Accenture for e-krona pilot but Switzerland Skeptical Of Central Bank Digital Currency

No alt text provided for this image
  1. Everywhere on the planet, dozens of companies — largely unregulated, little scrutinized — are logging the movements of tens of millions of people with mobile phones and storing the information in gigantic data files.

  2. The Worldwide Web of Chinese and Russian Information Controls and Baidu vs Google: 2019’s searches inside and outside China’s Great Firewall

  3. DreamQuark raises €14M for its AI software Brain and DataRobot to acquire data prep company Paxata

  4. More than 30% of consumers report owning a voice-activated speaker – more than triple the number who reported owning one over the three years PYMNTS has been tracking this – and nearly as many reported using it to make a purchase

  5. JPMorgan Employs Machine Learning For Expense Reports

No alt text provided for this image
  1. How Virtual and Augmented Reality Are Advancing Pediatric Care

  2. Dior Returns to Instagram AR with Virtual Makeup Effect to Promote Holiday Collection

  3. With the HOVER app, users are able to take smartphone images of a property and automatically receive accurate measurements, including the total living area down to the inch, as well as an AI-derived sketch of their home.

  4. China’s car parking problem spawns start-ups with big money backers as AI dominance grows

  5. UK Fintech Curve Expands Wearable Offering By Forming Partnerships With Garmin, Fitbit, & Sony’s Wena Pay


Looking for more?

Unicorn valuations are bad for your health -- the Prisoner's Dilemma of WeWork and Magic Leap; plus 20 key Fintech developments

Hi Fintech futurists --

In the long take this week, I look at why unicorn valuations can be destructive for the Fintech industry. While some companies do have incredible financial performance, much of the valuation premium in the private markets today is a negative Prisoner's Dilemma equilibrium, which leads to price compression, over-scaling, and problems with exit for everyone but the CEO. We look at this question through the lens of WeWork and Magic Leap, and the recent developments in the news about the economics of those companies.

The latest key updates on Fintech bundles, Crypto and Blockchain, Artificial Intelligence, and Augmented and Virtual Reality are below. Thanks for reading and let me know your thoughts by email or in the comments! Last but not least, these opinions are personal (or maybe made by a robot) and do not reflect any views of ConsenSys or other parties.


Long Take

Why are high valuations bad? You've heard me talk about how the trend of Fintech bundling, and the unicorn and decacorn valuations led by SoftBank and DST Global, are creating underlying weakness in the private Fintech markets. Of course, they are also creating price compression and consolidation in the public markets (e.g, Schwab/TD, Fiserv/First Data) across sub-sectors. But public companies are at least transparent and deeply analyzed. Private companies have beautiful websites, charismatic leaders, and impressive sounding investors. Often when you look under the hood, it's just a bunch of angry bees trying to find something to sting.

No alt text provided for this image

As Fintech venture investment continues to grow and the power laws of the web apply themselves to neobanks, roboadvisors and the rest, highly-valued private companies accumulate and consolidate. See: Now $500 million richer, $6B-valued Chime eyes acquisitions. According to Pitchbook, we have 140 or so multi billion dollar private companies since 2006 in play. Private means illiquid. Private means that the employees are frustrated that they have not sold any equity for a while. On the other side of the equation, selling that equity is taking longer and longer. The average time to an offering has gone from 3 years in 1996 to about 8 years in 2016. The amount raised in the private markets prior to IPO has similarly scaled from $12 million to $100 million. And that's clearly ignoring the economics of big platforms like Uber.

No alt text provided for this imageNo alt text provided for this image

It's taking so long that several Silicon Valley players (e.g., Slack, Spotify, Asana) are trying to avoid IPOs entirely, and just place a direct listing. The SEC is frustrating this attempt for the NYSE, helping defend investment banking business as a result, but the direction of gravity is inevitable. Initial Coin Offerings, Initial Exchange Offerings, Security Token Offerings, various blockchain-based bond issuances, and the rest are showing the world what direct security ownership will look like in the future. But that's not our point today. Our point is that over-valuation is actually bad business for the companies involved.

Let's anchor in WeWork and MagicLeap. Two articles this week shed light on what happened behind closed doors, though notably both articles are behind paywalls: Dented Reality: Magic Leap Sees Slow Sales, Steep Losses and The Money Men Who Enabled Adam Neumann and the WeWork Debacle.

The core Magic Leap story is that the company burned $50 million a month and raised over $2.5 billion. After completing some heavy and interesting R&D efforts, it failed on commercialization -- selling no more than 6,000 of its augmented reality headsets for about $3,000 per unit. As a result, Magic Leap just kicked off a major pivot from a consumer gaming company to an enterprise productivity company. In part, this is because the enterprise AR market has seen some chunky revenue, in the form of a $500 million Microsoft HoloLens contract with the US military. Consumer headsets have seen slower adoption, with about 10 million units per year in total, including VR rigs. Outside of PokemonGo, AR has struggled.

No alt text provided for this imageNo alt text provided for this image

Did it help Magic Leap to raise billions in capital, burn it down, and try to go all out in a market where there is no demand? You can see in the chart above that Consumer AR is not yet meaningful as a hardware category. It could end up being important for SnapChat, which already has a multimillion user footprint with built-in virality around content, fashion, and 3D rendered memes. It could be important for Apple or Google, which have a billion user footprint with hardware embedded across the entire world. Does $3 billion in venture funding really compare with a set of companies that print hundreds of billion in revenue by releasing the most innovative technology out there? All you've done with that much capital is prevented your acquirers from being able to afford you.

No alt text provided for this imageNo alt text provided for this image

What new thing is there to say about WeWork? The WSJ article gave a few really interesting data points about how WeWork was positioning its revenue growth for each fundraise. The reality is they were sinking deeper and deeper into an economic black hole with every dollar raised, as the high level financials below show. But the WSJ plotted these actuals against the projections in the investment documents. You can see the rosy "up and to the right" arrow just moving along every year. Even as the net loss increased to $2 billion on $2.5 billion in revenue, the only thing changing from a venture capital perspective was the starting point for the model.

No alt text provided for this imageNo alt text provided for this image

I can tell you with 100% certainty that every single entrepreneur trying to raise money does this exact trick. The main difference is that most people get to lose a few million, while Adam got to lose several billion and walk away with a $1.7 billion severance as compensation. At the end of it all, I don't really even think it was his "fault". Instead, this is the logical result of a blitz-scaling venture capital system, which creates a Prisoner's Dilemma environment for investors and operators both. The individuals in charge happen to represent the underlying qualities of the system in which they operate, but they are not special.

The machine is simple. The 2008 crisis has led to interest rates being artificially set to near zero for a decade. Technology and open data commoditized the returns in active public equity investing. Large insitutional investors, which normally hold meaningful allocations in both actively managed fixed income and equities, flocked to alternative investments in the form of venture capital and private equity. Private equity is a large asset class, and can generally absorb the capital, especially with cheap credit available. Venture capital is a small asset class and has ballooned, with too much money looking at the same "good" deals and bidding up valuations. Investors began to chase not operating performance, but each other -- engineering valuation increases between different vehicles held by the same owner, to create paper returns and motivate follow-on fundraising from markups.

No alt text provided for this imageNo alt text provided for this image

Normally, you would have a sequenced, smooth path of growth for a new company. It starts in the ideation phase, searching for the first customers and raising below $5 million at Seed stage, and up to $25 million for a Series A. That already can be seen as too rich. But it should be enough to get you to some sort of product-market fit. The further out you are on the frontier of new technology (e.g., AI, blockchain), the longer it will take you to find that repeatable customer. But generally, you shouldn't be raising $100 million checks for anything other than putting fuel into a rocket. Growth stage is where you operationalize scale, and build the right type of organization for a company that is generating profitability on the marginal customer. On the other side of this is just being an industry incumbent, and raising hundreds of millions through a public offering.

All this has been collapsed together in our current cycle, and massive companies like WeFox, Chime, Brex, Robinhood, Revolut, and the other various unicorns are just a symptom. They may have great underlying economics, or they may just have large user footprints -- it is hard to know. But regardless, venture investors try to pick winners earlier and earlier! And when they can't pick them through market competition, they try to create them through massive funding to lock up the market and cut price. These are the tools of a monopolist, with Peter Thiel as the closest thing to a prophet: Peter Thiel on How to Build a Monopoly. How much of Uber's success required a massive fundraising strategy and global scale? Probably all of it.

No alt text provided for this image

So what's the downside again? The issue is that we are in a negative Prisoner's Dilemma equilibrium. Take the diagram above. If any one firm takes SoftBank / DST Global investment and has the chance to blitzscale, its competitors are going to see price compression and are more likely to lose given a smaller war chest. As a founder/CEO of that target company, you probably want to take the crazy multi-billion dollar valuation both for your resume, and to cash out "a few dozen million" which seem insignificant in context. As the founder/CEO of the competitor, you have envy. So you go and find your own big check, and the industry becomes a lose-lose proposition.

The high valuations make it much less likely that company employees and smaller common shareholders see a positive liquidity event. The investment rounds come with a liquidity preference in the cap table, which postulates that for $2 billion of invested capital, $2 billion or more must be paid out first to investors on sale of the company. Debt is even more senior, which is why seeing Magic Leap collateralize its $280 million loan from JP Morgan with its intellectual property doesn't inspire confidence. All this math puts common shareholders deep underwater by the time it comes to being acquired or doing an IPO in the public markets. But that's a problem that nobody owns.

Maybe you can just do a direct listing for them!


Featured Interviews, Podcasts, and Conferences


Key Fintech Developments

I am tightening up this section by removing the short takes, and keeping in just the headlines and links. A lot of this stuff should be floating up into the Long Take anyway. Let me know if you dearly miss off-the-cuff musings, or if 20 links is too many links!

No alt text provided for this image
  1. BlackRock, Vanguard Among Fund Giants Flocking to China.

  2. Corporate venture capital deals hit new record as banks invest in fintech competitors.

  3. Wefox, the Berlin-based insurtech, raises $110M Series B extension at a $1.65B pre-money valuation.

  4. Robinhood Rolls Out Interest-Earning Bank Account.

  5. Brex, the start-up that lends to other start-ups, taps $200 million line from Credit Suisse.

No alt text provided for this image
  1. Twitter is Going Decentralized with Blockchain

  2. ING working on digital assets custody technology - sources

  3. China Construction Bank (CCB), the world’s second-biggest bank by operated assets, has officially launched its blockchain-based refactoring platform and Bank of China Uses Blockchain to Issue $2.8B Worth of Financial Bonds

  4. SEC green-lights $15 billion asset manager's bitcoin futures fund and Binance Futures hits a daily trading volume ATH of over $2.7 billion

  5. Illiquidity and Bank Run Risk in Defi

  6. The first part of Ethereum’s massive planned upgrade has been successfully activated on the main network.

No alt text provided for this image
  1. BoA’s Virtual Assistant Exceeds 10M Users; Launches New Features

  2. A "uniform loss ratio" test can eliminate bias in underwriting and open the way for truly individualized, AI-driven assessments of risk.

  3. China has four times as many surveillance cameras installed than the US, but is just behind America when it comes to number of CCTV cameras per capita

  4. As AI firms race to put autonomous taxis on the streets of China, the Post‘s Jane Zhang took a test ride using WeRide’s RoboTaxi service

No alt text provided for this image
  1. How Augmented Reality in E-Commerce Is Transforming Customer Experience.

  2. Magic Leap Pivots to Enterprise, Announces New Business-Focused Services, Slightly Modified Name.

  3. Snapchat Unveils Limited Edition Gucci Brand Spectacles With Insanely Bizarre Promo Video

  4. Nutanix and Hardis Team up on Supply Chain Transformation using machine vision in warehouses

  5. Trucking Payments Tools That Are In It For The Long Haul.


Looking for more?

Loading more posts…