How should investors think about factor capacity? Based on our research, there are three ways we can measure capacity, and all three suggest the scope for factor investing is massive relative to assets under management today.
As of October 31, 2019, the total US factor ETF assets—factor investing done through ETFs—were at $482 billion.1 Factor investing is one of the fastest growing areas of asset management, with approximately 78% of the aggregate growth in factor assets coming from net new flows (22% from appreciation).2
Chart 1: U.S. Factor ETF asset growth
Source: BlackRock, iShares Global Business Intelligence.
*Data YTD as of October 31, 2019
As factor ETF assets have climbed, a fair question is just how large factor strategies can become before the amount of capital begins to weigh on returns?
The answer: A LOT.
The question then turns to how investors should think about factor capacity? Based on our research there are three ways we can measure the capacity of factor-based strategies:3
Transaction costs
When you invest in a factor fund, you ultimately invest in a selection of stocks and will incur transaction costs. As more investors put money into the fund, transaction costs increase. At what point do the historically observed premiums in factor strategies become eroded by trading frictions?
To answer this question, we estimate factor capacity using a proprietary transaction cost model, built up from fixed costs and market impact for each individual security. Not surprisingly, the factor with the smallest capacity is momentum: the high standalone volatility and turnover of the momentum strategy leads to relatively high transaction costs—but even here, the capacity of momentum at a five-day trading horizon is $324 billion. Currently, there is only $12 billion in this strategy, which is less than 4% of capacity! Conversely, the capacity for minimum volatility is $6.8 trillion compared to only $88 billion being invested today, and the capacity for multi-factor combinations (value, quality, momentum, and low size) is in excess of $1.5 trillion!
Chart 2: Capacity for a Five-Day Trading Horizon AUM and Capacity ($B)
For illustrative purposes only. Capacity calculations as of June 14, 2016: See Capacity of Smart Beta Strategies: A Transaction Cost Perspective: Ratcliffe, Miranda and Ang (2016). *AUM data based on BlackRock, iShares Global Business Intelligence data as of October 31, 2019. All factors are representing the factor universe in smart beta strategies as described in the paper.
New money vs existing money
To provide some context on the size and scope of factors, l et’s first review the total size of the broader investment universe.
The S&P 500 is the world’s most-tracked index. As of December 31, 2018, the total market cap of the S&P 500 was $22.1 trillion. According to S&P Dow Jones Annual Survey of Assets, active US equity strategies account for $5.6 trillion of that market cap (25%), while passive market cap index strategies account for $3.6 trillion (16%). (Security holdings by other investors such as hedge funds, sovereign wealth funds, pension funds, insurance companies and other financial institutions, and individuals account for the remaining $12.9 trillion.) Comparatively, total US factor strategies would account for less than 1% of this stock ownership, with minimum volatility strategies less than .004%. The scope for factors is enormous by comparison!
Chart 3: Size of US Factor Strategies Relative to S&P 500 Index
Source: Dow Jones Inc., S&P Global 2019 Investor Fact Book
*BlackRock data as of October 31, 2019
Traditional active mutual funds have significant factor exposures: they are underweight minimum volatility, underweight value, and tend to hold momentum stocks. Another way to measure capacity is to estimate how much money would be required for these active funds to neutralize their under- or over-weights.
Chart 4: Aggregate factor exposures for all active mutual funds
Aggregated Factor Exposures for All Active Mutual Funds
Source: Barra, BlackRock, Morningstar, MSCI, Thomson Reuters as of 3/31/2016
Take minimum volatility as an example. Active funds tend to be underweight minimum volatility because these funds need to take risk, relative to benchmarks, to generate returns. As such, they are underweight to low risk stocks. In order for these active funds to eliminate their underweight to minimum volatility they would need to collectively invest $650 billion in minimum volatility! (Compared to just $88 billion in total US minimum volatility AUM today.)
Factor premiums and invested capital in equilibrium
A final way to quantify the capacity of factor strategies is to jointly estimate factor premiums and invested capital. For the nerds who might appreciate it, the estimates of factor capacity using transaction costs or active fund holdings are partial equilibrium calculations. But in equilibrium, prices and quantities are determined together.
Just like a seesaw, large flows into a set of securities will push their prices upward, and their future expected returns decrease. Similarly, large flows out of subset of stocks will send prices falling, and push up risk premiums. With an appropriate economic model, we can estimate the extreme shifts in capital from one strategy to another required to eliminate the historically observed factor premiums.
We estimate factor premiums when all the capital from the lowest 30% of expensive stocks (growth) is moved to the top 30% of the cheapest stocks (value). That is, we buy the factor with the high return (value) and sell the stocks on the other side. This is a hypothetical flow of $13 trillion. This flow would reduce the value premium by approximately half( from 3.8% to 1.9% per year),but importantly the value premium still exists.
There are some factors where even with these extreme movements in capital, the factor premiums barely move. For example, doing this same exercise for the minimum volatility premium, we find the premium drops slightly from 6.1% to 5.4% per year. Modeling the same exercise for momentum strategies, the momentum premium shrinks from 7.4% to 5.9%.
Chart 5: Capital transfer required to attenuate the factor risk premiums
Source: Ang, Hogan and Shores (2018), Factor Risk Premiums and Invested Capital: Calculations with Stochastic Discount Factors, Journal of Asset Management, Vol. 19, No. 3, pp.145-155. ges/faculty/ken.french/data_library.html.
Factor capacity is enormous
The large capacity of factor strategies—estimated from transaction costs, examining factor loadings of active funds, and using economic models to link returns and capital invested—reflects the enduring nature of their economic underpinnings. It’s the economic rationales of a reward for bearing risk, structural impediments, or investors’ behavioral biases that explain why the long-run capacity of factors is large, and why factors will continue to be broad and persistent drivers of returns for a long time to come.
But large long-term factor capacity doesn’t mean that factors don’t vary over time—and tilting factors around strategic allocations can be, for certain investors, a way to generate additional incremental returns.
Andrew Ang, PhD
Head of Factors, Sustainable and Solutions for BlackRock Systematic
Andrew Ang, PhD, Managing Director, is Head of Factors, Sustainable and Solutions for BlackRock Systematic. He also serves as Senior Advisor to BlackRock Retirement ...
