Week 47 Day 1: What Rebalancing Is and Why It Matters

Example: You start January with $600,000 in stocks and $400,000 in bonds (60/40). By December, stocks have risen 20% and bonds 2%. You now have $720,000 stocks and $408,000 bonds -- that is 63.8/36.2. To rebalance, you sell $33,600 of stocks and buy $33,600 of bonds, returning to 60/40 with your new total of $1,128,000. Without rebalancing, a good year for stocks quietly increases your risk.

Rebalancing serves two purposes: risk control and return enhancement. Risk control is intuitive -- you prevent your portfolio from becoming riskier than you intended as one asset class outperforms. Return enhancement is less obvious but equally important. By selling high and buying low systematically, rebalancing captures a 'diversification return' or 'rebalancing bonus.' Bernstein (1996) estimated this bonus at 0.2-0.5% annually for a diversified portfolio. The mechanism is simple: asset classes mean-revert over long periods. When stocks have a great year, their expected forward returns are slightly lower (higher valuations). When bonds lag, their yields have risen (higher expected returns). Rebalancing forces you to shift money toward the asset with better forward prospects. The psychological challenge is real: rebalancing requires you to sell your winners and buy your losers. Every instinct screams against it. That is precisely why it works -- it is systematically contrarian. The alternative -- never rebalancing -- means your portfolio allocation is determined by whatever asset happened to perform best recently. That is momentum investing by default, which works until it does not (see: 2000, 2008).

The theoretical foundation for rebalancing rests on Markowitz (1952) mean-variance optimization, which shows that a fixed-weight portfolio on the efficient frontier offers the best risk-adjusted return for a given level of risk tolerance. Any drift from the target weights moves the portfolio off the efficient frontier, reducing the Sharpe ratio. Empirical work by Plaxco and Arnott (2002) examined rebalancing across multiple asset classes from 1968 to 2001 and found that disciplined rebalancing added approximately 0.4% per year in returns while simultaneously reducing portfolio volatility. Booth and Fama (1992) formalized the concept of 'diversification return' -- the difference between the weighted average return of the components and the return of the rebalanced portfolio -- showing it is always positive for a portfolio of volatile, imperfectly correlated assets that is regularly rebalanced. The magnitude of the diversification return increases with (a) higher volatility of the components, (b) lower correlation between components, and (c) more frequent rebalancing up to a point. However, Kitces (2015) notes that the rebalancing bonus is not free alpha -- it is compensation for the systematic contrarian behavior (buying low, selling high) that rebalancing enforces, which requires emotional discipline most investors lack.