Original compilation: BlockTurbo
Original compilation: BlockTurbo
Portfolio management is hard.
Of the 26,000 stocks traded since 1926, only 1,000 have accounted for all the profits in stocks over the past 100 years. Only 86 stocks (0.33%) contributed half of these gains. Its hard for individual stock pickers to win. Even the best portfolio managers underperform the market 80% of the time.
Portfolio management is even more difficult when you are trying to manage a portfolio of start-up projects with unknown regulatory, event and technology risks. Of the roughly 22,000 cryptocurrencies in existence, there are fewer than 86 options that are likely to deliver long-term returns for the asset class.
This is the border, friends. But despite the risks, there may also be some nuggets to be reaped for those willing to take the risk.
first level title
Portfolio Theory and Crypto Applications
Modern Portfolio Theory (Modern Portfolio Theory, MPT) is basically the basis of all portfolio management. The theory allows investors to build a portfolio with an expected level of return for an assumed level of risk. The greater the risk, the higher the reward; the less risk, the lower the reward. In theory it is.
The MPT guides investors in assessing the risk (measured by standard deviation), return (measured by some time series of historical average returns) of each asset class they wish to include in their portfolio, and the correlation between them. Using this data, we can build a weighted portfolio and estimate its historical expected performance under a range of conditions. We can then combine assets with greater confidence to construct the optimal portfolio based on our risk parameters. Spend any amount of time in any introductory finance course and these principles will come up time and time again.
MPT provides a useful lens for constructing long-term portfolios. With a long history of how these asset classes have performed across a range of scenarios and how they interact, we can guide long-term allocation decisions.
But when trying to build a crypto portfolio, the problem is apparent.
For most traditional assets, the long term is years or more, if not decades. Cryptocurrencies have a short history of performance, high volatility, and correlations are largely erratic.
Using MPT in traditional portfolio management, allocators attempt to address expected returns to meet or exceed their benchmarks. If youre a pension fund, your benchmark might be some fixed number (i.e. 7%). If youre an endowment, its probably a goal above some set payout rate (ie 4% per year). If youre an individual investor or a fund, it might be a simple benchmark (ie 60% stocks/40% bonds). These allocators move along the risk/reward spectrum until they can build a portfolio containing a mix of assets that has historically been likely to help them achieve their desired return with the lowest level of perceived risk.
In addition to serving as a tool for measuring investment performance, benchmarks provide easy access for passive investors seeking market exposure. If planning to invest passively, benchmarks and portfolios are the same thing. For example, if you plan to buy the SP 500, buy SPY, the index that represents the benchmark. Your portfolio and benchmark are essentially the same.
first level title
BTC and ETH Performance in a Modern Portfolio
First, analyze the contribution of BTC and ETH to the portfolio. Today, most managers or investors exposed to cryptocurrencies judge their performance based on these two assets alone. Investors considering an allocation to cryptocurrencies also naturally turn to these two most famous assets.
This is not surprising since both BTC and ETH have very high initial returns. Bitcoin has returned more than 300,000% since 2009, while Ethereum has returned more than 5,000% since 2014. The SP 500 gained about 260% and 140% in those two periods, respectively. The Sharpe ratio, a measure of risk-adjusted performance, has consistently averaged well above 1 for both assets.
The Sharpe ratio penalizes the up and down volatility of an investment significantly. This is not a huge problem in traditional asset management, but it is a meaningful problem in cryptocurrency. The Sortino ratio adjusts the Sharpe ratio to account for downward bias only. Historically, BTC and ETH have performed on par with or outperformed stocks and bonds on this metric. Since 2019, Bitcoin’s Sortino Ratio has outperformed the SP 500 66% of the time, while Ethereum has outperformed the index 67% of the time.
Outside observers could argue that much of this impressive risk-adjusted performance is due to the very low base these assets are based on. Almost any measure of asset performance will look good if any asset ranges from $1 or less to thousands of dollars. Such a performance is unlikely to be repeated in the future.
Thats a fair criticism. However, starting in 2018, we can evaluate datasets that are starting to become more stable. If we use this time series to add BTC and ETH to a standard 60/40 portfolio, the benefits are still evident. The higher the configuration of BTC and ETH, the better the performance.
Although cryptocurrencies have been volatile throughout 2018, 2020, and 2022, adding BTC and ETH to traditional portfolios has provided high returns of around 10%, at the expense of very little increase in portfolio volatility. Traditional portfolio analysis is then extended to how these assets perform in terms of:
Periods of rising/falling interest rates
periods of rising/falling inflation
expansion/contraction economy
This is the basis for future analysis as the encrypted sample size increases.
Risk and return are important, of course, but so is the correlation between assets in a portfolio. Portfolios are combined so all assets dont move in the same direction at the same time. The correlation of Bitcoin and Ethereum with traditional assets has been erratic. Correlations typically rise in tandem with risk assets, which appears to reduce the overall gains that BTC and ETH can bring to a broader portfolio. The overall correlation of daily returns to the SP 500 over the past two years is about 0.8 for Bitcoin and 0.6 for Ethereum. This suggests that the diversification benefits of holding these assets are small. This correlation has trended down over time, from a five-year streak of above 90%, opening the door for further diversification gains.
secondary title
Best Portfolio Analysis for BTC and ETH
In addition to adding simple weights of BTC and ETH to traditional portfolios, we can also use portfolio analysis to find the optimal portfolio. For this analysis, since 2018, we have optimized a series of portfolios along the efficient frontier using the returns, risks, and correlations of these assets.
Monte Carlo analysis simulated market conditions for 1 million trials to find the optimal risk-adjusted portfolio. The analysis concluded that the best portfolios are bonds (62%) and ETH (27%).
According to analytical data, BTC and ETH offer empirically and academically proven advantages to portfolios. As data becomes more widely disseminated and understood by the public, more investors are likely to continue to increase their allocations to these assets. The next step for these investors will be to consider cryptocurrency-specific portfolios.
A naive market-cap weighted benchmark would formulate a portfolio of approximately 40% BTC and 20% ETH. Some investors may stick to the two main crypto assets, but many will choose to invest more broadly to take advantage of the asset class.
secondary title
Benchmarking
Today, there are a range of existing cryptocurrency benchmarks in addition to Bitcoin and Ethereum. Assets for these strategies vary widely, even for funds that serve as broad market benchmarks. The table below outlines some of the more popular indices. Here, the terms index and benchmark are used interchangeably because they are usually the same. For example, the SP 500 is also the most popular US stock market benchmark.
Compared to traditional markets, the benchmark selection is significantly less. There are a limited number of dedicated cryptocurrency benchmarks, and the U.S. stock market alone has thousands of them. Even more worrisome is the composition and subjectivity of these indices. Invest in an index and you expect broad exposure to that asset class for what you pay: not just two assets. The average cost of the U.S. broad market index is as low as 0.03%. Today, some cryptocurrency-specific fees, such as custody fees, are higher, but that certainly doesnt explain the 70x difference in costs. Cost is not the only issue. Investors leave performance on the table (discussed in the next section).
The best option is the blockchain-native BED index run by Index Coop. It maintains broad exposure across many assets at very low fees. Unfortunately, the AUM of the BED index is around $1 million, while the AUM of the Bitwise 10 is over $400 million. Self-custody still appears to be a barrier to adoption, although we may see more adoption as Coinbase and others add BED.
In addition to the concentration issues of broad industry benchmarks, there is a range of inherent biases. If the benchmark is not market-cap weighted, the selection of assets is extremely subjective, and assets are often added at peak speculation. Some indices hold small-cap tokens (i.e. Enjin, Sandbox, Axie), others do not; some indices hold stablecoins, others do not. Some place caps on individual assets, resulting in an underweight BTC/ETH relative to the market.
In the stock world, similar decision points have distinct differences. We see tags like Small Cap, Equal Weight, Growth, Value, Technology, Energy, etc. An emerging industry of crypto industry-specific benchmarks, but the problem with broad benchmarks is worse here.
Other investors, such as venture capital, hedge funds, and private equity, who allocate to the leading edge of emerging industries, suffer from similar benchmark and index imperfections due to the volatility of the underlying series and the frequent influx of new investments. There is no best solution to these problems - only time to better evaluate the data series.
Proper use of passive indices is critical to an investors long-term success. In traditional finance, 90% of active large-cap managers underperformed the index over a 10-year period. Surprisingly, the situation is even worse for small caps (91%), where you would think there would be more opportunities to find rough diamonds. Passive options are the best way to capture the performance of a given market.
first level title
secondary title
Crypto Portfolio Optimization
What would an optimal risk-adjusted portfolio (i.e., Sharpe-optimized) look like if one analyzed a market portfolio of top crypto assets with at least two years of history? The best portfolio is (obviously) a mix of Loopring and Avalanche. An unconstrained analysis optimized for the Sharpe ratio will obviously not give us as useful results for a crypto portfolio as a similar analysis in traditional markets. This is likely due to the limited dataset and skewness curves of the assets.
To remove the effect of volatility, the top assets are optimized here using the Sortino ratio. The results become more useful, showing that optimal risk-adjusted portfolios are closer to the equal-weight approach.
Portfolios that are overweight to small-cap assets are known to provide risk-adjusted returns to stocks at the expense of slightly higher volatility. Access to smaller assets is especially important to take advantage of emerging technologies.
It’s worth noting that the closer to equal weighting your crypto portfolio is, the better your risk-adjusted performance will be. Crypto performance really reflects risk, with small wins driving overall performance.
To better optimize liquidity, transaction costs, and portfolio management time, at least market-weighted Bitcoin and Ethereum are more realistic. Therefore, at a minimum, the cryptocurrency market cap weights for Bitcoin and Ethereum highlighted earlier should be our maximum allocation. Thereafter, the rest of the portfolio can be maximized. The allocator might consider something closer to that portfolio or the lowest cost, broadest spread passive index. The resulting portfolio is a composite of multiple sectors.
Utilizing an equal weighting approach or a modified bitcoin and ethereum portfolio offers simple improvements that should result in less concentration, more upside potential and more diversified crypto market exposure. As new assets emerge, they should be considered for inclusion in a portfolio based on a range of factors, including potential total addressable market capitalization. The emerging trends we highlighted in 2023, such as decentralized social, L2 scaling, and DePIN, come to mind.
first level title
secondary title
RebalancingRebalancing
The time frame for rebalancing can have a meaningful impact on absolute and risk-adjusted portfolio performance, especially when investing in volatile asset classes. A more frequent rebalancing strategy dampens the impact on volatility and returns. Academic research shows that these effects are even more meaningful when an asset class is less correlated with its portfolio counterpart.
As outlined by Bitwise, the difference in returns from regular rebalancing is huge for both cumulative and risk-adjusted returns. An annual rebalancing plan appears to maximize returns while greatly reducing drawdowns and volatility. Regardless of the benchmark, a rigorous rebalancing strategy is paramount.
secondary title
type analysis
In the world of traditional assets, we also analyze different types of stocks. Are they growth or value? Are they large caps or small caps? These distinctions help investors allocate based on where they think they may be in the market cycle. Value stocks tend to outperform growth stocks during periods of rising interest rates, tightening monetary conditions and slowing economic growth. Large-cap stocks tend to outperform small-cap stocks in similar situations.
While we certainly dont currently exist, what will cryptographic classification look like in the future? New classification systems are currently being developed, but we can begin to classify major cryptoassets according to their emerging characteristics. This is especially important for dedicated crypto asset managers who lack the discretion to allocate meaningful portions of their portfolios outside of the asset class.
secondary title
Strategic Asset Allocation
We try to map crypto assets to the traditional stock space. Different stock sectors have been shown time and again to perform better under certain market conditions. A business cycle framework can help understand how to make better tactical allocations. For example, financials and energy tend to benefit in a rising interest rate environment.
secondary title
Liquidity and Market Depth
The US stock market operates 37.5 hours per week (excluding pre-market and after-hours trading). The cryptocurrency market never stops. A deep market is one in which a trader can trade assets in large volumes without fear of moving the market and receiving increasingly unfavorable prices as orders are executed. Liquidity and volatility have a strong negative correlation in traditional markets. Not only are crypto markets roughly 50 times smaller than U.S. equities, but their trading volumes are spread over a wider time frame. This has resulted in periods of lack of depth and illiquidity, even for major assets.
The lack of liquidity in the crypto market has exacerbated volatility. The lesson for investors is to monitor their exposures more closely during times of worst liquidity, especially if leveraged. Volumes were highest at the open in the U.S. and slightly higher again at the Asian open. This U.S.-centric trading pattern mirrors traditional markets, but given the adoption in Asia-Pacific compared to the U.S., cryptocurrencies are somewhat surprising.
Volatility peaks around the U.S. market close when liquidity is at its lowest. Realized volatility nearly doubled between early and late U.S. trading.
Volatility breeds alpha opportunities. Excess return opportunities in the equity space are measured in basis points; in cryptocurrencies, we measure it in percentages. Its not uncommon to see assets move by double-digit percentages within a day for little reason. What drives these violent movements? While its hard to point to an obvious reason, liquidity is certainly an important factor. Understanding market depth and liquidation levels is key to outperformance.
technical analysis
technical analysis
Technical analysis is one of the most widely debated topics in finance. In its most basic form, technical analysis looks for repeatable patterns in charts of stocks, bonds, currencies, and more. Technicians try to identify patterns and trends and try to position themselves to profit from them. This sits alongside the fundamental analysis that most people do when analyzing a company or other investment, looking at the underlying quantifiable factors that might drive it, such as earnings and capital investment. The basic assumption of technical analysis is that prices move in trends; history repeats itself and patterns can be observed and exploited.
For some, its boon; for others, its rubbish, just people drawing lines on a chart. There are academic studies that refute it. There are academic studies pointing to it being excellent in certain modes. The easiest way to think about technical analysis is that while it may or may not work, most investors are aware of major technical patterns. So, in a sense, some technological configurations have become a kind of self-fulfilling prophecy. Technical analysis is certainly something to watch out for. Especially in the field of encryption.
With most crypto assets lacking fundamental support, technical analysis has actually proven somewhat useful. A long-only machine learning model trained on past BTC returns was able to outperform a simple buy-and-hold strategy with fewer drawdowns.
These technical phenomena are often arbitraged in traditional markets, but seem to persist in cryptocurrencies to some extent. The reason for this persistence may be due to the fact that there are fewer capital chasing opportunities in this space compared to other major markets. This is especially true after hedge funds are phased out in 2022, and opportunities for capital deployment remain fairly limited. Cryptocurrency liquidity among major assets remains very thin compared to the stock market. Spot trading volume on exchanges in December was $357 billion; Nasdaq alone traded that amount in about two days.
Summarize
Summarize
Original link
