FORECASTS & TRENDS E-LETTER
by Spencer Wright
September 10, 2024
IN THIS ISSUE:
The Good Old Days to The Present
The Benefits of Computerized Markets
The Knight Capital Disaster and Drawbacks of Computerized Markets
Final Thoughts
The Good Old Days to The Present
When you think of stock or commodity market trading you probably picture large crowds of traders shouting and gesturing at each other. That was known as the open outcry system. It is still what I picture and was the norm when I got into this business 30 years ago. Open outcry has been on the decline for 15 years and is now dead as disco. The evolution of electronic trading roughly follows the course below.
The Early Years (1980s-1990s): Algorithmic trading began as a means to automate trading strategies, primarily focusing on market making and arbitrage opportunities. Early algorithms were simple, rules-based systems that executed trades based on predefined conditions.
Development of Complex Strategies (2000s): As computing power and data storage increased, algorithmic trading evolved to incorporate more sophisticated strategies, such as mean reversion, momentum, and statistical arbitrage. These complex algorithms enabled traders to capture a wider range of market inefficiencies.
Advent of High-Frequency Trading (HFT) (2005-2010): HFT emerged as a distinct category of algorithmic trading, characterized by extremely short-term trading horizons (milliseconds to seconds) and high-speed execution. HFT algorithms exploited market microstructure inefficiencies, such as order book imbalances and latency advantages.
Regulatory Responses and Market Impact (2010-2015): As HFT grew in popularity, regulatory bodies began to scrutinize its practices, leading to increased transparency and oversight. Market volatility and flash crashes, partly attributed to HFT, prompted concerns about the stability of financial markets.
Advances in Machine Learning and Artificial Intelligence (2015-Present): The integration of machine learning and artificial intelligence (AI) into algorithmic trading has enabled the development of more sophisticated and adaptive strategies. These AI-powered algorithms can learn from market data, adjust to changing conditions, and improve their performance over time.
You may be wondering what percentage of markets are traded or influenced by algorithms. Consider this infographic:
Keep in mind that this is on a global scale and reflects mainly North American, European and Asian markets. Currently India’s market is about 40% algorithmic but it is increasing quickly.
The Benefits of Computerized Markets
As the capacity for electronic trading evolved and market algorithms grew in prevalence and capability, there was definite unease. Regulators and investors alike voiced concerns that removing people from the trading and market making equation entirely was a bad idea. But there were many benefits that markets and investors experienced from technology advancement and trading automation.
The main benefits of algorithmic trading include:
Speed: Algorithms can execute trades in fractions of a second. This speed enables traders to capitalize on market opportunities quickly and accurately.
Reduced Costs: Algorithmic trading minimizes transaction costs by optimizing order execution and reducing market impact.
Scalability: Algorithms can handle large volumes of trades, making them ideal for institutional investors and high-frequency traders.
Accuracy: Algorithmic trading eliminates human emotions and biases, ensuring that trades are executed based on predefined rules and mathematical models.
Backtesting: Algorithms can be tested on historical data, allowing traders to evaluate their performance and refine their strategies before implementing them in live markets.
Execution at Best Prices: Algorithmic trading enables traders to execute trades at the best available prices, improving overall performance.
Improved Liquidity: Algorithmic trading can create liquidity by automatically executing trades and providing market makers with opportunities to profit from order flow. (This has become a very big deal.)
Systematic Approach: Algorithmic trading provides a more systematic approach to active trading, replacing intuition and instinct with mathematical models and predefined rules.
Reduced Human Error: Algorithms eliminate the risk of human error, such as incorrect order entry or failed trades.
Increased Transparency: Algorithmic trading provides a clear and transparent record of trades, which allows regulators to better follow the action.
These factors are remarkable. Today, markets are for the most part one penny wide. That leads to better order fills and general price improvement on trades for retail investors as well as zero commission stock and ETF trading.
These efficiencies have also pleased regulators as trading is more transparent and the possibility of human error is greatly reduced. Of course, the are potential downsides to algorithmic trading.
The Knight Capital Disaster and The Drawbacks of Computerized Markets
“On August 1, 2012, Knight Capital fell on its sword. It experienced a software glitch that literally bankrupted the company. Between 9:30 am and 10:15 am EST, the employees of Knight capital watched in disbelief and scrambled to figure out what went wrong as the company acquired massive long and short positions, largely concentrated in 154 stocks, totaling 397 million shares and $7.65 billion. At 10:15, the kill switch was flipped, stopping the company’s trading operations for the day. By early afternoon, many of Knight Capital’s employees had already sent out resumes, expecting to be unemployed by the end of the week.” (Source: spechbranch.com)
I urge you to read this narrative on the Knight Capital disaster. It is an excellent and straightforward account. The event lasted 44 minutes and cost Knight $440M. Soon after they were purchased by a competitor and ceased to exist.
While you can argue that this was at its root a human error, it gave pause to many regarding the potential perils of algorithmic trading. Here are a few:
Dependence on technology: Runaway algorithms can result in insurmountable losses for traders. (See above) Many blame algorithms for the flash crash of 2010.
Front-running: Is a significant regulatory concern. Quant funds use advanced algorithms to detect impending large orders by institutional investors, enabling traders to profit from front-running these securities.
Error-prone: One glitch in the program can trigger many orders, leading to a dramatic rise or fall in stock prices, causing panic and volatility.
Markets have also become more volatile because of algorithmic trading. What used to take markets days or weeks to resolve is now done in a matter of hours or days. Daily one and two percent moves are now commonplace.
Final Thoughts
There is no going back to the good old days of wide markets, sluggish liquidity, thick trading fees and overall inefficiency. Markets are forward-looking. Having been along for nearly the entirety of this transformation I find that I am thinking a great deal about the impact of widespread AI implementation in the markets. What will that look like and how will it change markets and trading? I don’t know but we are most definitely going to find out.
And now some parting fun. I had MS Copilot render the following image using the prompt “AI stock market.”
Thanks for reading,
[Spencer]
Spencer Wright is an investment advisor with Halbert Wealth Management, Inc. and a regular contributor to Forecasts & Trends. He has been with HWM for over twenty-five years and serves on the Due Diligence Committee and the Investment Committee. His experience in domestic and international investments gives him valuable insight to those markets.