Dimensional Models: Low Turnover plus Active Implementation

The Dimensional Wealth Models, launched on March 31, 2020, are strategic asset allocation models composed of Dimensional funds and are designed to put rigorous theoretical and empirical research to work across an investor’s portfolio.

By design, the models aim to have low turnover in top-level model weights, allowing for cost- and tax-efficient implementation.¹

Does the lack of high turnover at the top level mean lack of active implementation at the fund level? Not at all. The Dimensional funds that make up the Wealth Models are managed daily to incorporate real-time market information and systematically pursue higher expected returns while managing risks and controlling costs.

Consider the fixed income sleeve of the Core Wealth Models. Since launch at March-end 2020, the target weights of the fixed income components have not changed. Yet fixed income markets have changed quite a bit, as illustrated by rising interest rates and a period still ongoing during which the US yield curve has inverted. Consistent with a daily implementation approach, the funds that compose these models have changed to incorporate updated information in global fixed income markets.

In particular, when yield curves are upwardly sloped and term spreads are wide, Dimensional may extend the average duration of a fund due to larger expected term premiums. When yield curves are flat or inverted and term spreads are narrow, Dimensional may shorten the duration due to lower expected term premiums.

We see this pattern in Exhibit 1, where we focus on the 60/40 and 80/20 Core Wealth Models to illustrate. (These models hold 60% and 80% in equities, and 40% and 20% in fixed income, respectively.) Overall, the average duration of the 80/20 allocation is higher than that of the 60/40 allocation, consistent with higher risk tolerance and growth-of-wealth objectives typical of investors in more equity-heavy models.² We plot the weighted average duration of the fixed income components of the Core 60/40 and 80/20 Wealth Models against the US term spread, as defined by the difference in yields of US government intermediate bonds vs. US government 1–3 year bonds, from March 31, 2020, through June 30, 2023. The US term spread rose from 12 basis points (bps) up to 49 bps in March 2021, when the yield curve was steeply upward sloping, and then trended downward as the curve flattened and inverted, landing at –48 bps as of June 30, 2023.

The average duration of the Wealth Models has followed a similar pattern, as the underlying fixed income funds dynamically adjusted their duration exposure over time. For both models, the average duration increased as term spreads widened; as term spreads narrowed, the duration shortened.

Similarly, Dimensional can dynamically vary the credit exposure of the fixed income strategies to capture the reliable relation between credit spreads and credit premiums.

Exhibit 2 plots the weighted average portion of the fixed income sleeve of the models in A/BBB rated bonds against the US credit spread, as measured by the difference in yield between US intermediate A/BBB rated bonds vs. US intermediate AAA/AA rated bonds, from March 31, 2020, through June 30, 2023. Following a spike in credit spreads in March 2020 around the COVID-19 crisis, both the credit spread and weight in A/BBB rated bonds generally declined until mid-2021, and then slowly increased until mid-2022.

The changes in fixed income characteristics over time reflect an active approach to implementation, incorporating real-time information in markets to target higher expected returns. And, because this approach to implementation happens within the model components themselves, we do not need to adjust fund allocations at the model level.

Dimensional’s systematic active approach also applies to equities, where we target higher expected returns by pursuing the size, value, and profitability premiums. Research has not found compelling evidence for timing equity premiums, so we maintain a steady and balanced focus on the premiums in our models.³ But staying consistently focused on the reliable drivers of higher expected returns requires incremental rebalancing within the equity strategies.

As the characteristics of a stock change, its expected return also changes. Our daily implementation process allows us to take this into consideration. One way to see that is through the strategies’ sector allocations, which change over time depending on how the size, value, and profitability characteristics of the different sectors within a country evolve. If many securities in a sector exhibit higher expected return characteristics (lower market capitalization, lower relative prices, and higher profitability), our emphasis on these securities will result in a limited overweight in that sector relative to its natural market weight in the size-eligible universe. And, if many stocks in a sector exhibit low expected return characteristics relative to the market, we will underweight securities in that sector. The rebalancing across sectors and industries happens seamlessly within the equity portfolios and does not require any model-level turnover.

By way of example, consider the two largest component fund positions in the Core Plus Wealth Models: US High Relative Profitability Portfolio and US Vector Equity Portfolio. Exhibit 3 presents the sector weights in both portfolios relative to the market at the time of model inception, March 31, 2020, and, more recently, as of June 30, 2023. Sector weights in the US High Relative Profitability Portfolio are shown relative to the Russell 1000 Index; for the US Vector Equity Portfolio, the Russell 3000 Index is the comparable universe.

The US High Relative Profitability Portfolio is shown in Panel A. In March 2020, financials were less profitable relative to other companies in other sectors, which led to an 8.5-percentage-point underweight of that sector in the high-profitability strategy. As the profitability ranking of financials in the cross section improved over time, the underweight dropped to 2.5 percentage points as of June 30, 2023.

Within the US Vector Equity Portfolio in Panel B, the relative weight to financials fell from an overweight of 12.4 percentage points to an overweight of 8.0 percentage points, as the book-to-market ratio of financials decreased (financial companies exhibited higher relative prices) over the period. The changes in sector allocations happened all within the funds themselves and did not require any rebalancing at the model level.

These outcomes highlight the value of integrating a strategic approach to asset allocation at the model level with a daily implementation approach at the fund level. For investors, our daily process helps to ensure our models maintain a consistent focus on securities with higher expected returns and add value through daily consideration of real-time market information. For advisors, our strategic approach makes model management simpler, reducing the level of trading needed to track the models and making them easier to explain to investors.