Robo-advisors provide investors automated financial advice with varying levels of sophistication and degrees of individual tailoring. In their December 2019 book chapter entitled “Robo-advising”, Francesco D’Acunto and Alberto Rossi catalog the main features of robo-advising with respect to personalization, discretion, involvement and human interaction. They consider robo-advisors designed to assist short-term and medium-term (active) trading and those designed to guide long-term (passive) investment/accumulation for retirement. They review prior research on effects of robo-advisors regarding investment choices and performance. Based on the body of information on robo-advising, they conclude that:
- Four key aspects of robo-advising are:
- Personalization – Robo-advisors typically elicit client demographics such as income, investment horizon, risk tolerance, job security and existing portfolio allocations. They then assign the client to a category and propose a matched investment strategy. They generally do not account for non-financial investments or upcoming major expenses. Some robo-advisors offer special features such as tax loss optimization.
- Involvement of the investor – Robo-advisors for active trading typically ask clients to approve each recommended trade, re-optimizing after rejected trades. Robo-advisors for passive investing (robo-managers) typically trade automatically on behalf of clients.
- Strategy discretion – Some robo-advisors let clients modify proposed portfolio weights (with or without pre-set guardrails) and insert stocks and other assets not in the robo-advisor universe, re-optimizing after such client interventions.
- Human interaction – Many robo-advisors are purely automated to maintain low operating costs. Some catering to older, wealthier clients use algorithms for portfolio allocation but assign human advisors for client sign-up and responses to questions.
- Robo-advisors typically apply mean-variance optimization to maximize Sharpe ratio, aiming to: prevent overexposure to risky assets; maintain adequate diversification; and, suppress destructive behavioral biases.
- Some robo-advisors more comprehensively seek to guide client saving for retirement by collecting personal spending/savings data and generating an personal balance sheet, with occasional “nudge” messages about saving. Some offer clients crowdsourced information about average spending, assets, debts and net worth of others with similar demographics.
- Prior research on effects of robo-advisors finds that:
- For stock traders who are initially undiversified, robo-advisors generally increase diversification, reduce portfolio volatility and produce slightly higher average return. For those already diversified, robo-advisors typically do not improve portfolio performance and may hurt average return by increasing number of rebalancing trades.
- Overall, robo-advisors improve risk-adjusted performance by reducing portfolio risk. They accomplish this improvement by lowering allocations to money market funds, individual stocks and active equity funds and raising allocations to bond funds, low-cost equity index funds and to international equity and bond funds.
- Robo-advisors benefit inexperienced clients who have large cash holdings and trade frequently. They also benefit clients with few mutual fund holdings and those holding high-fee active mutual funds.
- Robo-advisors boost client Sharpe ratios by an average 10%, mostly by reducing idiosyncratic risk and portfolio volatility.
In summary, the body of information about robo-advisors indicates that they mostly apply conventional portfolio construction methods that are beneficial to unsophisticated investors.
Cautions regarding conclusions include:
- The authors do not address robo-advisor fees.
- Evidence often does not support belief that mean-variance optimization is the best way to set portfolio allocations.
For other research summaries addressing robo-advising, see:
