The Limits of DeFi for Financial Inclusion

1. DeFi and crypto-asset market trends in ASEAN

The ASEAN region is home to the fastest growing internet user base, with a large proportion of the population being young urban citizens keen to adopt new technologies. This demographic group is most comfortable with using crypto-assets and participating in decentralised finance markets. This could partly explain the elevated levels of activity in markets for decentralised finance, comprising crypto-assets, including stablecoins, and DeFi protocol activity.

This chapter analyses DeFi and crypto-asset market activity in AMS in the period 2020-22 and examines trends, patterns and possible underlying drivers of peaks and troughs in activity in ASEAN.

1.1. Decentralised finance market trends in ASEAN

Activity in decentralised finance markets has been recorded in all ASEAN Member States, although the level of activity differs significantly across the countries, in line with the varying economic activity and diverse technological developments in the region (Figure ‎1.1). ASEAN economies are embracing digital technologies to varying degrees especially when it comes to institutional and infrastructural readiness, digital skills, and digital payments activities as evidenced by the ASEAN Digital Integration Index (ASEAN, 2021[1]).

Figure ‎1.1. Diverse level of decentralised finance activity in ASEAN

Note: Measured by crypto-asset flows per capita. Aggregate flows for the period 01/2020 to 05/2023, Average Population for 2020 and 2021.

Source: OECD based on Chainalysis flows and World Bank population data.

Crypto-asset flows towards ASEAN over the period 2020-22 follow a pattern that tracks to some extent the price evolution of Bitcoin (Figure ‎1.2). Indicatively, increased activity is recorded around October 2021, when Bitcoin reached a historical high of USD 61 525 (18 October, based on data from Chainalysis). This is not surprising as Bitcoin’s price has been a major driver of overall decentralised finance activity globally. Feedback loops exist between Bitcoin, the price of which has driven much of the activity in the entire crypto-asset market, and the evolution of DeFi measured by the total value of such crypto-assets that is locked into smart contracts of DeFi protocols (OECD, 2022[2]).

Activity in ASEAN peaked in H2 2021, in line with global trends, and started subsiding in early 2022 at the onset of the crypto-asset market downturn (known as ‘crypto winter’) and in line with overall global trends in these markets (OECD, 2022[3]). Interestingly, crypto-asset flows towards ASEAN almost tripled in the aftermath of the Terra UST de-peg on 7 May 2023 (see Box ‎1.1). A total of USD 17bn of flows to ASEAN were recorded on 8 May 2022, compared to USD 6bn a week before. This could possibly be attributed to withdrawals from Terra-connected Anchor protocol in the wake of the UST de-peg, and to massive withdrawals from among similar DeFi protocols, among other things, by ASEAN-based investors.

Increased activity of decentralised finance markets in times when crypto-asset valuations are high is an indication of speculative forces driving these markets. Indeed, speculation and the potential for high returns are considered as the main drivers of investor interest and participation in this space (OECD, 2022[4]). There are no fundamental drivers of the surge in crypto-asset prices in what seems to be a market largely driven by speculation, given the very high returns that can be achieved due to the massive volatility of crypto-assets (excluding stablecoins), and the fear of missed returns (so-called fear of missing out or ‘FOMO’). These forces are intensified given the extensive recycling of profits from some crypto-asset activities to others for example from mainstream crypto, such as Bitcoin, to DeFi protocols. The reflexive character of crypto-assets further intensifies these trends of common pattern in levels of activity asset types and regions.

Figure ‎1.2. Crypto-asset activity in ASEAN

Note: Measured by total flows of crypto-assets in ASEAN. On a monthly basis, includes Bitcoin, Ether and Altcoin crypto-assets, and stablecoins.

Source: OECD based on Chainalysis data.

Box ‎1.1. The Terra-LUNA implosion and its impact on retail investors

The rise and fall of Terra’s so-called stablecoin (UST) in a short time span is a useful case study exposing the risks involved in decentralised finance markets. UST issuance grew exponentially from 2% to 10% market share in one year with a USD 18 bn issuance volume as of April 2022.

On 7 May 2022, UST broke its peg with the US Dollar and suffered a run that resulted in significant losses for UST holders (UST declined from USD 0.99 to less than USD 0.13 in the course of five days). Some UST retail investors lost a significant part of their investment without any recourse for compensation. For example, in the case of Terra UST’s implosion, 280 000 investors reportedly lost their investments (news1.kr, 2022[5]), as a result of whichnearly 4 400 retail investors have formed an association called ‘the UST Restitution Group in pursuit of compensation from the Terra founder (FT, 2022[6]). Such investors appear to have been drawn into UST by the unsustainably high yields offered by protocols connected to UST (e.g. through the Anchor protocol) and with little understanding of the circular and reflexive character of these crypto-assets, which had no fundamental value.

The subsequent de-pegging of the largest so-called stablecoin at the time, Tether’s USDT, which was reportedly backed by holdings of commercial paper, did not have an immediate observable contagion effect on short-term debt markets. However, it clearly highlighted the high interconnectedness within crypto-asset markets and gave rise to consideration of potential risks stemming from transmission channels to traditional markets, that did not materialise in the end. The UST implosion also highlighted the limits to creating unbacked private money, as well as the central role of trust and confidence in the market for crypto-assets. This is analogical to the importance that trust plays in traditional financial markets, where a sharp loss of confidence can affect the crypto-assets ecosystem. The failure of Terra’s dominant stablecoin UST created contagion and had a domino effect in the broader crypto-asset market leading to the failure and bankruptcy of other crypto-asset firms such as Celsius and Three Arrows, ultimately culminating in the FTX collapse in November 2022.

Source: (OECD, 2022[3]; Nassr, 2022[7]).

1.2. Breakdown of decentralised finance activity by ASEAN Member State

Thailand recorded the highest crypto-asset flows per capita in the period 2020-22, followed by Malaysia and Viet Nam. Important crypto-asset flows per capita were also recorded in Brunei Darussalam, the Philippines and Lao PDR, and to a smaller extent in Indonesia, Cambodia and Singapore (Figure ‎1.1).

In absolute terms, Viet Nam has received the largest amount of cumulative crypto-asset inflows for the 2020-22 period examined, with a total of USD 0.19tn of crypto-inflows recorded in that period, followed by Thailand with USD 0.18tn (Figure ‎1.3). Singapore, the Philippines and Indonesia also recorded significant amounts of crypto inflows (USD 0.12tn, 0.11tn, and 0.09tn respectively). On the contrary, there has been little activity in Brunei Darussalam and Myanmar. Cambodia and Lao PDR recorded small absolute activity that, however, has been non-negligible on a per capital level (Figure ‎1.3).

Viet Nam, Thailand and the Philippines were among the top 10 crypto-asset adopters in 2022 according to some industry analysis (Chainalysis, 2022[8]). More than a third of the population in Malaysia, Viet Nam and Thailand have reported having used or owned crypto-assets based on an industry survey (Statista, 2023[9]).

Figure ‎1.3. Significant levels of crypto-asset activity in many ASEAN Member States

Note: Period from January 2020 to May 2023, average GDP 2020-21.

Source: OECD based on Chainalysis flows and World Bank population data.

Figure ‎1.4. Share of respondents in AMS who have used or owned crypto-assets from 2019 to 2023

Note: Based on a sample of 2 000–12 000 respondents per country.

Source: Statista Global Consumer Survey.

1.2.1. The role of ‘play-to-earn’ blockchain-based gaming

Blockchain-based ’play-to-earn’ gaming applications that involve crypto-assets may have fuelled the surge in new adopters in these three AMS. “Play-to-earn” gaming involves non-fungible tokens (NFTs) and relies on the use of Ethereum-based crypto-assets for the gaming economy. Viet Nam is the headquarter of ex-unicorn company Axie Infinity, a leading online gaming company, and its developing team Sky Mavis.1 Thailand and the Philippines are also known as countries with increased blockchain-based gaming activity.

In-game rewards in crypto-assets have indeed lured in many young individuals who aimed at supplementing their income in some AMS. According to one estimate, out of 2.8m daily active users, 65% are estimated to come from South-east Asia with the majority being Filipinos (as of February 2022) (Knowledge Partners, 2022[10]). Rewards in crypto-assets within such systems depended on the value of such unbacked crypto-assets2 and could therefore not be considered as a viable supplement to participants’ income or conducive to financial inclusion in these countries. Anecdotal evidence suggests that participants from the region were left indebted as they could not earn enough to cover the borrowed funds for their initial participation in the platform (Time, 2022[11]).

Crypto-assets involved in blockchain-based gaming are exchanged in crypto-asset exchanges and carry the same risks as any crypto-asset, and as such, risks taken by participants involved are likely to outweigh any benefit of additional income for users. This includes risk of total loss of holdings through exploits or hacks, as was the case with the USD 540m hack that Axie Infinity suffered in March 2022 by hackers linked to North Korea (Elliptic, 2022[12]). Depending on the jurisdiction, they may also involve non-compliant provision of financial services or activity outside the regulatory perimeter, exposing participants to material risks (OECD, 2022[3]).

Increased crypto-asset activity in Malaysia and (to a lesser extent) Thailand can also be attributed to crypto-asset mining activity. Malaysia is one of the top 9 countries with crypto-mining activity and Thailand ranks 2^nd of 10 ASEAN countries (see Figure ‎1.6). Crypto-mining provides a flow of newly-created Bitcoin in particular, and incentivises trading activity and the creation of an ecosystem around the mining activity.

Figure ‎1.5. Share of respondents of Play-to-earn gaming adoption in ASEAN as of 2022

Source: (Finder, 2022[13]).

1.2.2. Crypto-mining activity in ASEAN

Evidence of crypto-asset mining activity has been recorded in 7 AMS, with the largest part of such activity happening in Malaysia and some non-negligible activity also taking place in Thailand. In fact, Malaysia is listed as one of the 9 countries with the largest Bitcoin mining activity as estimated by the Cambridge Bitcoin Electricity Consumption Index, accounting for 4.2% of global Bitcoin mining (Figure ‎1.6). Thailand accounts for 0.6% of global Bitcoin mining activity, and there is minor mining activity estimated to take place in Indonesia, Viet Nam, Philippines and Cambodia.

The main motive for crypto-asset mining is the associated financial rewards: mining revenues consist of the sum of transaction fees paid to miners, plus the market value of newly minted coins awarded to miners who successfully compete to record transactions to the blockchain. Holding all else equal, Bitcoin mining firms’ profit margins increase when the price of the underlying crypto-asset increases, and therefore mining activity – and the corresponding energy consumption – is greater when the price of the underlying crypto-asset increases (OECD, 2022[14]). That explains the increase in activity in periods of crypto-asset market exuberance.

Figure ‎1.6. Bitcoin mining activity

Source: Cambridge Bitcoin Electricity Consumption Index.

The People’s Republic of China (hereafter ‘China’) has previously been dominating crypto-asset mining with more than half of total Bitcoin mining occurring in the country until January 2021, when crypto-asset trading and mining was banned over environmental concerns, given the heavy environmental footprint of this activity (OECD, 2022[14]). The ban led to relocation of mining activities to other parts of the world, such as the US and Kazakhstan, places that offered affordable energy sources. Although it could be presumed that part of the activity in Malaysia may have been supported by such relocations given geographical proximity, the increase in mining activity in Malaysia after the imposition of China’s ban has been almost negligible. Additionally, it is estimated that miners in China moved towards masking their real locations in order to avoid being identified as operating in a country where restrictions apply (e.g. through IP VPN tools) (OECD, 2022[14]). Reasons related to energy supply, associated energy costs as well as local legislation may also partly contribute to such activity in ASEAN (Fortune.my, 2018[15]).

It should be highlighted, however, that there are important data gaps around crypto-asset mining activity that remain to be overcome. The importance of addressing such data challenge is high given the environmental impact of such activity and the need for robust and reliable data to support policy assessments and possible action (OECD, 2022[14]). It should also be noted that miners engage in activities that would be illegal in traditional markets such as front-running and sandwich trades, which leads to additional costs for retail investors (Auer, Frost and Vidal, 2022[16]).

Box ‎1.2. Crypto-asset mining and the negative environmental footprint of crypto-asset activity

The Bitcoin blockchain and other DLTs that validate transactions using proof-of-work consensus mechanisms (PoW) rely on the energy-intensive process of ‘mining’.1 Mining involves the use of aggregated computing power, and powerful computing hardware, to participate in the mechanism that determines what is recorded on a blockchain. As an incentive for consuming CPU time and electricity to support the network, mining participants are rewarded with new coins and transaction fees.

The use of energy-intensive transaction validation using PoW consensus mechanisms and the corresponding carbon footprint creates climate transition risks for participants in these markets. It can undermine the progress towards the climate transition and may not be correctly priced in, especially given the possible lack of awareness of such footprints by some of the holders and the lack of robust data about the measurement of such impact.

Given the carbon footprint and associated climate transition risks of certain digital assets, policy consideration and potential policy action is warranted. Outright banning of crypto-asset mining activity may not be the most effective solution to address these issues, given the risks of displacing or concealing such activity instead of curbing it, as was the case with the China ban of crypto-asset mining in 2021. This ban resulted in (a) the relocation of miners to other geographies with the same net effect for the globe; and (b) the concealing of activities still operating in China through the use of technical means.

In the absence of cross-border co-ordination, the negative externalities are being transferred from one geography to another, while the overall negative impact to the environment and societies remains the same.

1. In September 2022, the Ethereum blockchain on which the second largest crypto-asset (Ether or ETH) is based, transitioned from PoW to proof-of-stake (PoS) through a process known as ‘the Merge’, resulting in a 99% reduction of its environmental footprint.

Source: (OECD, 2022[14]).

1.3. Types of crypto-assets in ASEAN

Decentralised finance activity in ASEAN has been dominated by Bitcoin and stablecoins (Figure ‎1.7). In terms of unbacked crypto-assets, Ether is also heavily used, although to a lesser extent than Bitcoin. This follows global trends of preferences for these two mainstreamed crypto-assets, particularly when it comes to holdings by professional or institutional investors, either directly or through regulated products referencing Bitcoin (OECD, 2022[4]). As Bitcoin is the crypto-asset with the longest history and renown, it may inspire more investor confidence for having stood the test of time in the short life of crypto-markets. Both Bitcoin and Ether have gone through several cycles with deep bear markets, possibly creating the expectation by retail investors of continued patterns of cycle recovery.

Figure ‎1.7. Breakdown of crypto-asset activity in ASEAN by type of asset

Note: includes Bitcoin, Ether, Altcoin crypto-assets and stablecoins. Stablecoins consist of BUSD, DAI, TUSD, USDC, USDP, USDT_BTC, USDT_ETH, USDT_TRX and Altcoin crypto-assets consist of DOGE, FTT, LTC, TRX, WBTC, WETH, XRP.

Source: OECD based on Chainalysis data.

1.4. The importance of stablecoins in the ASEAN ecosystem

The second largest category of assets in the ASEAN market is so-called stablecoins, which have historically constituted at least 50% of total flows in ASEAN throughout the period 2020-22 (Figure ‎1.7). Although initially designed by the industry to serve as a means of payment, with a stated ambition to be used widely by retail users to pay for goods and services and by corporations in the context of supply chain payments, stablecoins are used in their vast majority by crypto-asset investors. So-called stablecoins have been heavily used to facilitate trading, lending and borrowing and other transactions involving digital assets since the early days of development of crypto-asset markets. For example, stablecoins are used to move between crypto-assets or crypto-exchanges (e.g. lending, borrowing, collateral pledged on DeFi lending protocols, liquidity mining in DeFi), or to reduce their exposure to crypto-assets without having to convert their holdings to fiat and/or exit the decentralised finance space. Given that crypto-asset exchanges do not allow trading across all possible crypto-asset pairs, stablecoins’ primary role has been that of a medium of exchange within these markets, similar to a euro-dollar instrument in non-US exchanges, and especially given their utility across different blockchains. Anecdotal evidence suggests that stablecoins in emerging market economies are also being used as a way to hedge against weak or volatile domestic currencies.

The proportion of stablecoins as a percentage of total ASEAN flows has increased significantly since the crypto winter in early 2022, which is in line with the global trend in the behaviour of crypto-asset investors wishing to protect themselves from the increased volatility of unbacked crypto-assets (e.g. Bitcoin). As such, investors have been shifting from unbacked crypto-assets towards stablecoins to reduce their exposure to the downturn, given the perception of relative stability of these instruments (that is, relative to unbacked crypto-assets). Stablecoins have therefore allowed crypto-asset holders to protect against mainstream crypto-asset volatility without having to exit the decentralised finance market or covert their holdings into fiat. Some crypto-asset holders could be intentionally avoiding the use of national currencies or traditional financial institutions to exit their investment in order to conceal illegal activity and/or evade taxation of earnings, among other reasons.

Figure ‎1.8. Claimed use cases of so-called stablecoin arrangements

Stablecoins are an indispensable ingredient and one of the foundations of DeFi protocols, and the increased DeFi activity in ASEAN may partly explain the increased proportion of stablecoins in overall crypto-asset activity in the region. They serve as collateral in lending protocols; are used to transfer funds from one protocol to another and are supplied to liquidity pools for swaps in Decentralised Exchanges (DEXs) and Automated Market Makers (AMMs) (OECD, 2022[4]). Also, as stablecoins are less volatile relative to other crypto-assets, holders of stablecoins may seek additional yields by investing in DeFi protocol activity, especially if they cannot exit these markets and convert to fiat – inter alia due to taxation constraints.

Figure ‎1.9. Cumulative stablecoin flows in ASEAN from 2020 to 2022

Note: Includes the following stablecoins: BUSD, DAI, TUSD, USDC, USDP, USDT_BTC, USDT_ETH, USDT_TRX. Based on average GDP for 2020 -2021 based on World Bank data. Direct flows are defined as a transfer of assets where the source and destination services are both counterparties to the transfer. Indirect flows denote a transfer of assets where at least one counterparty to the transfer is a self-hosted entity but where the ultimate source or destination of the transfer is described.

Source: OECD based on Chainalysis data.

1.5. DeFi protocol activity in Asia

DeFi protocol activity has been one of the fastest-growing3 parts of the crypto-asset and wider decentralised finance ecosystem since the summer of 2020 (also known as the “DeFi summer”). Total value of crypto-assets locked in DeFi protocols on the Ethereum blockchain stood at 2.4bn in July 2020, and reached USD 108.8bn at the peak of the market in November 2021 (Figure ‎1.10). DeFi activity “tracks” the trend of mainstream crypto-assets given massive interconnectedness, dependencies and feedback loops between DeFi protocols and crypto-assets (OECD, 2022[3]).

Figure ‎1.10. DeFi and crypto-asset market evolution

Source: DeFiLIama, from January 2020 to June 2023; Refinitive from January 2020 to June 2023.

Figure ‎1.11. DeFi ASEAN flows

Source: OECD based on Chainalysis data.

DeFi protocol activity globally has been concentrated in DeFi lending and borrowing protocols and in decentralised exchanges (DEXs) and Automated Market Makers (AMMs) (OECD, 2022[2]). Activity in DeFi lending and borrowing protocols has been driven to a large extent by unrestricted leverage opportunities on the basis of crypto-assets. Crypto-assets are pledged as collateral in DeFi lending protocols in exchange for borrowed crypto-assets that can in turn be pledged as collateral on different protocols, building leverage in the system (OECD, 2022[4]). Most DeFi protocols have been historically built on the Ethereum blockchain given smart contract functionalities provided by the ERC-20 Token Standard. This could explain the important Ether flows towards DeFi globally and in ASEAN, too (Figure ‎1.12).

Figure ‎1.12. Inflows of Ether to DeFi by country in 2022

Source: OECD calculations based on Chainalysis data.

DEXs and AMMs4 are disintermediated, peer-to-peer on-chain platforms where traders can swap crypto-assets for other crypto-assets - but not for fiat, as in most centralised crypto-exchanges.5 Instead of order books, DEXs rely on liquidity pools where investors (liquidity providers) lock crypto-assets in exchange for rewards. Trading volumes in DEXs tend to follow the wider crypto-asset market volume trends, and particularly that of the Bitcoin, knowing that most crypto-asset prices tend to move in sync. Empirical evidence suggests that the most important crypto-assets are moderately positively correlated with each other over time, when daily and weekly returns are considered, and that the correlation strength increases significantly during downturns (Lahajnar and Rožanec, 2020[17]).

DEXs are an increasingly important part of the DeFi market, with growing market share as compared to centralised crypto-asset exchanges (CEXs), particularly following the recent failures of multi-function crypto-asset intermediaries and in spite of similar vulnerabilities around operational and cyber security risks. The vast majority of crypto-asset transaction activity had been happening at centralised exchanges for crypto-assets (CEXs) until early 2021, when DEX trading volumes started to become a meaningful part of the total activity, growing to almost 20% of the total trading activity as of H2 2022. Interestingly, DEXs have experienced a surge in trading volumes in the aftermath of the FTX failure, the Silicon Valley Bank and other major crises related to the crypto-asset market. This could be attributed to investors’ lack of confidence and fear of losing access to their crypto-assets, given that DEXs involve self-custody of assets by the investors themselves without a third-party custodian.

Figure ‎1.13. Increasing part of crypto-asset trading occurring in DEXs

Note: CEX is defined as crypto-to-crypto exchanges and DEX is as decentralised exchanges

Source: OECD based on Chainalysis.

A depiction of estimated aggregate crypto-asset flows across continents indicates important flows from Asian countries to the DeFi space both on an aggregate basis for the period 2022 - H1 2023 (Figure ‎1.14) and for each of the years in that period (see Annex). This estimate, however, includes flows from large markets such as China, Japan and Korea. It should be noted that DeFi protocols have in most cases no defined jurisdiction and allow participation on a pseudonymous basis without KYC onboarding. It is therefore difficult to obtain accurate statistics of geographic breakdown of DeFi activity by geography.

Figure ‎1.14. Aggregate net flows across geographies from 2020 to 2023

Note: Based on net flows measured in USD. DeFi represents smart contracts that facilitate financial intermediation of cryptocurrencies, for example lending or crowdfunding. Decentralised exchanges are included in the exchanges category.

Source: OECD based on Chainalysis data.

The prevailing crypto-assets in ASEAN are Bitcoin, Ether and the major so-called stablecoins USDT (Tether) and USDC (Circle), in line with the global trend of dominance of mainstream crypto-assets (Figure ‎1.15). The importance of Bitcoin flows in ASEAN has been decreasing over the years; Bitcoin flows constituted almost a third of all flows in 2020 and dropped to 15% of total as of 2022. Flows of Ether followed an opposite trend, accounting for 13.5% of total flows in 2020 and increasing to a third of total flows in 2021, and 20.0% of flows in 2022, perhaps driven also by the rise of DeFi protocols that were primarily based on the Ethereum.

USDT has been the prevailing so-called stablecoin used, accounting for a third of total flows to ASEAN, followed by USDC with 17% of total as of 2022. Stablecoins are issued on different chains, and in order to have a view on the usage of a particular stablecoin, the sum of the same stablecoin issued in different chains has to be considered (e.g. USDT issued on the Ethereum plus Tron blockchain are considered in this analysis). The pervasiveness of stablecoins is underlined by their multiple usage in crypto-asset systems (see Section ‎2 ‎1.4).

Figure ‎1.15. Yearly Crypto-asset flows to ASEAN from 2022 to 2022

Note: Direct flows: a transfer of assets where the source and destination services are both counterparties to the transfer.

Indirect flows: a transfer of assets where at least one counterparty to the transfer is a self-hosted entity but where the ultimate source or destination of the transfer is described.

Assets: BTC, BUSD, DAI, DOGE, ETH, FTT, LTC, TRX, TUSD, USDC, USDP, USDT_BTC, USDT_ETH, USDT_TRX, WBTC, WETH, XRP.

Source: OECD based on Chainalysis data.

Notes

← 1.  AXS token market value reached USD 10bn at its peak and amounted to roughly USD 600 million as of December 2023.

← 2.  In some cases, the dependence of the value of such crypto-assets on the inflow of new clients was a characteristic shared with Ponzi schemes.

← 3.  And following the crypto-asset market downturn, one of the fastest-contracting parts of the wider decentralised finance market.

← 4.  e.g. Uniswap. AMMs are DEXs that pool liquidity from users and price the assets within a pool using algorithms. For the purposes of this report, DEX and AMM terms are used interchangeably.

← 5.  e.g. Coinbase.