Effects of Maker-Taker pricing models on global Bitcoin markets
2 April 2020
Underpinning global exchanges and their bottom line is the maker-taker fee model. At the crux of this fee structure is rewarding liquidity provider, the maker, with lower fees or rebates, while charging a higher fee on the taker – the investors who ‘take’ liquidity away from the exchange instantly.
The maker-taker model was introduced into equity markets in 1997 and has been the prevailing fee structure since, under the economic theory that it would tighten the gap between bid and ask prices and, at the same time increasing liquidity on an exchange.
But there has been concern behind maker fee rebates. Chief Investment Officer of Yale University called rebates “kickbacks.” The Financial Times posits that the maker-taker model is “one of the most controversial elements of the US equity market.” Chief Strategy Officer at Bloomberg calls the maker-taker model a “prisoner’s dilemma.”
Retail brokerages Charles Schwab, Scottrade, TD Ameritrade, Fidelity, E* Trade, Edward Jones, and Morgan Stanley, were under investigation for potentially taking advantage of the rebate system against their clients.
And in a move to protect self-interest, the New York Stock Exchange (NYSE) and the Nasdaq began court proceedings by submitting two 400-page petitions last year against the US Securities and Exchange Commission (SEC) to block a pilot program that could see the end of exchange rebates.
But why all the controversy, and how does this affect cryptocurrency markets? ZUBR takes a look behind the economic theory, incentives, and the real effects behind asymmetric fee’s on exchanges.
• While maker rebates increase liquidity initially, the benefits diminish as more high-frequency traders compete against each other reducing profitability and order book size. This leads to less effective markets.
• Price discovery becomes an obvious problem as liquidity providers account for their net trade after rebates. This skews the trading spreads, and the real value across exchanges differs primarily due to fee structures.
• Maker rebates are akin to marketing tactics employed by exchanges to win over liquidity from competitors. The net revenue earned by exchanges is the same as the taker is burdened with the higher cost of trading.
• Maker rebates force passive investors to trade more aggressively in order to access liquidity as spreads are artificially narrowed.
• The US Securities & Exchange Commission has questioned the economic theory and incentives of maker rebates and is piloting a program that could possibly end the model. This would mark a different future for exchanges and trading as investors who 'take' liquidity would be better protected from routing practices that forfeit 'best execution' price.
The first trading system to introduce maker-taker rebates was Island ECN in 1997. The economic theory behind the incentive of rebates was to bring in liquidity to exchanges and reward makers for taking on the risk of patiently waiting for an order to fill.
This was a departure from the traditional fee model where both, buyer and seller, paid a small fee. The change in the structure had massive implications for exchanges and markets worldwide.
Traders and brokers became more inclined to trade on exchanges offering rebates, and the new model triggered exchanges to adopt the rebate model to stay competitive.
But as more embraced the new fee structure, new problems came to the fore. Regulators had to step in and put a cap on maker-taker incentives as exchanges began to increase rebates and hike taker fees only a year into the new paradigm. Fees on exchanges between maker and taker diverged significantly to win over liquidity to their venue.
“Maker-taker models seem more attractive to makers and reduce apparent spreads so it’s better marketing to display things that way, much like displaying prices with tax not included.”
Accounting for Reality
The primary driver behind the maker-taker fee was the assumption that makers having incentives to post liquidity would lead to tighter bid-ask spreads. To the model’s credit, by every observable metric, this has become a fact.
However, as traders managed their strategies across exchanges, it became more evident to researchers and regulators that the narrow spreads were an artificial accounting exercise. Liquidity providers would ultimately account for their final payout from a trade in direct relation to their rebate.
A joint research paper by USC, Georgetown, and Carnegie Melon concluded that “maker-taker pricing has not changed net spreads, but has decreased quoted spreads.”
A study by the University of Southern California similarly concluded that by” narrowing quoted bid-ask spreads, maker-taker pricing had introduced a transparency problem into the markets. Quoted prices do not reflect net prices.”
And the SEC found that “in the absence of maker-taker fees, the quoted prices would reflect more closely actual net economic prices. Consequently, if quoted prices more closely reflected actual net prices, then market participants may be more willing to reach across the market and take liquidity at the quoted price rather than seek to avoid an execution as taker in a maker-taker environment and consequently lock the market.”
Exchange Balance at Expense of Takers
Asymmetric fees may seem to be a fair market structure rewarding the liquidity providers. However, it should be noted that this is not at the expense of the exchange, but at the investors who are the takers being charged the higher fees. A letter penned to the SEC by RBC Capital Markets states that the maker-taker model “may be increasingly disadvantaging long-term investors who are interested in fundamental value.”
This fee dynamic forces passive market participants to trade more aggressively in order to access liquidity that is “often fleeting” under the maker-taker model which leads to another problem – accurate pricing and actual volume across fragmented markets with the variation in fees that aren’t baked into the quoted price.
Automated Trading Implications
As ZUBR previously showed in its last research on trading dynamics, global markets move together regardless of timezone. This indicates that cryptocurrency trading has become very automated and catering to a more sophisticated crowd.
While initially, this is a net positive as market effectiveness would increase, market execution and liquidity begin to decrease as more market participants compete on both the demand and supply side that would see their positions closed.
The Maker-taker model only exasperates the problem further, a position even the SEC (see table below) has taken as it pilots the controversial program that could put an end to rebates. Rebates' purpose, if not it’s the primary goal, is to incentivize market makers to profit by collecting on the discounts that are burdened by investors who wish to access liquidity.
But the opposite effect is likely to be occurring as spreads artificially narrow to the minimum and makers account for their net trade after rebates.
The Negative Effects of Maker-Taker Rebates
The length of exchange queues has increased, making it less likely for other market participants to provide passive liquidity with optimal execution.
Passive market participants are forced to trade more aggressively to access liquidity.
Liquidity is often fleeting – that is, it may disappear in times of market stress, particularly for less liquid (see charts below).
Myriad order types, rebates, and fees proliferate as trading venues seek to capture market share, resulting in excessive fragmentation of order flow.
Maker-taker proponents have asserted that markets will see an increase in trading volume due to posted liquidity. However, data shows multiple problems currently facing cryptocurrency markets due to its nature of high volatility.
Recent events highlight that growth in trading volume does not indicate an increase in liquidity available on markets, nor do maker-taker rebates keep spreads close during high volatility.
In 2020, up until March 12, just before markets began experiencing high volatility, BitMEX spreads went from an average of 0.28% to a whopping 4% (see chart 1).
Adding to the problem were prices that completely decoupled from major spot exchange prices, highlighting real stress in liquidity despite the exchange's maker rebate model (see chart 2).
Bitcoin: % Spread between Bid and Ask increased over 20-Fold during high-volatilty
Bitcoin: Prices differed as high as 22.8% on BitMEX vs Coinbase on March 13
As cryptocurrency markets develop more sophisticated tools and exchanges, the industry will need to increase its efficiency to address an asset-class that has limited supply and to date remains highly volatile. Although the prevailing model of maker rebates continues to be the norm, regulators are challenging the notion as market strains are not ironed out by incentivizing liquidity providers.
In essence, rebates prove to be more about market share grabs than about market effectiveness and efficiency. This leads to deeper problems in real price discovery that would allow for 'best execution' practices.
The growth of the market will need more market participants from makers and large brokers who can execute trades to the benefit of the investor. However, with rebate models, investors could be in the cross-hairs of conflicts of interest as brokers route their trades to exchanges with the highest rebates rather than the best price.
© 2020 ZUBR. All rights reserved.