- Longer investment horizons and a reliance on appraisals may mean that private real estate does not react to shocks as rapidly as public markets. However, this does not mean the asset class is immune to those shocks.
- As the COVID-19 pandemic continues to take its human toll and disrupt global economies, many real estate investors have been seeking to understand what impact the crisis could have on their portfolios.
- We used MSCI’s Valuation Scenario Model to show how discounted-cash-flow models could be used to explore the impact the crisis could have on asset values. In the hypothetical scenarios we created, asset values declined between 13% and 37% highlighting how sensitive the outcomes can be to changes in assumptions.
The COVID-19 pandemic has already had a notable impact on public equity markets, with the MSCI ACWI Index declining by just over 20% for the first quarter of 2020. By contrast, the impact on private markets such as real estate has been harder to establish. As a private asset class with lease structures and investment-hold periods that typically extend over multiple years, as well as a reliance on relatively infrequent appraisals, it has historically taken longer for adjustments to play out.1 However, real estate has not historically been immune to growth shocks. Discounted-cash-flow (DCF) scenarios may help investors better understand the potential sensitivity of their portfolios to those shocks.
Defining and comparing discounted-cash-flow scenarios
To demonstrate this approach, we define several hypothetical scenarios in the MSCI Valuation Scenario Model and compare them against the result of a hypothetical base case. The MSCI Valuation Scenario Model is a DCF-based tool. It derives cash flows based on growth expectations over short, medium and long horizons, which follow an auto-regressive process with mean-reverting features and discounts forecast cash flows with the expected risk-free rate and risk premium. By comparing valuations from scenarios with varying projections for growth and discount rates, the model analyzes the impact of growth shocks and discount-rate shocks on valuations.
Much of the media’s discussion to date has focused on the negotiations taking place between tenants and landlords and the potential for landlords to grant tenants relief in the form of rent deferral or rental holidays. For that reason, we have built our scenarios around these possibilities.
- In the base case, cash flows are assumed to grow at 3% per year and are discounted with a risk-free rate of 1% and a risk premium of 6%.
- In the first hypothetical scenario, we model a rent deferral where 95% of cash flows from the first six months are deferred. The risk-free rate falls to 50 basis points (bps), but the risk premium increases to 7%.
- In the second hypothetical scenario, we model a rent holiday with the same 95% reduction in cash flows over the first six months but the foregone income is not recovered. The risk-free rate falls 50 bps, but the risk premium increases to 7.5%.
- The final hypothetical scenario is the same as the rent holiday, but we model a more prolonged slowdown by reducing medium-term growth to 2% and increasing the risk premium to 8%.
Compared to the base case, under the rent-deferral scenario, the model shows asset values falling 13%. In the rent-holiday scenario, the model shows asset values fall 24% and in the prolonged-slowdown scenario, according to the model, the implied asset value falls by 37%. The model also calculates a cash-flow impact and a discount-rate impact, with the former applying the base-case discount rate to the hypothetical scenarios’ cash flows and the latter applying the scenarios’ discount rate to the base-case cash flows.
In the three hypothetical scenarios we’ve modeled in this analysis, the impact of the changes to cash flows on their own could result in asset-value declines between 2% and 13%. The changes to the risk-free rate and risk premiums in isolation implied a reduction in asset values of 12% to 28%.
Comparing the hypothetical scenarios to the base case shows the potential impact on asset values
|Scenario||Total||Cash-flow impact||Discount-rate impact||Interaction|
The cash-flow impact and discount-rate impact are not linear, so they will not always sum to the total. This difference is captured in the interaction term. Source: MSCI Valuation Scenario Model.
One thing to note in this analysis is that, while much of the public discussion to date has focused on the potential short-term disruption to income, the cash-flow impacts from the rent-deferral and rent-holiday scenarios, as modeled, are relatively small. Reductions to longer-term cash-flow expectations or changes to the discount rate had a much bigger impact on the results. This illustrates why some may want to consider what effects these variables could have on portfolios, in addition to the immediate disruptions to near-term cash flows.
With the situation evolving rapidly, there is still much uncertainty about how the crisis will impact the cash flows of real estate assets and financial-market conditions.
The above scenarios are not meant to be predictions, but to illustrate how investors can use tools like the MSCI Valuation Scenario Model to explore how sensitive a portfolio might be to changes in assumptions about growth and discounting. This sort of approach could be applied to the overall portfolio, but also to different segments within a portfolio. For example, the office exposure could be analyzed separately to the retail exposure, or the analysis could be applied to individual assets. In defining scenarios for this kind of analysis, investors can draw on their own internal underwriting assumptions, but the use of historical market data may also be helpful. It could be used to either define scenarios based on previous shocks or contextualize the assumptions applied to the current crisis.2
1For example, in 21 national market corrections we explored, the median peak-to-trough timing was 3.5 years. Reid, B. “What’s the downside in real estate?” MSCI Blog, Oct. 4, 2019.
2For an example of how historical data can be used to provide context to scenarios, see: Reid, B. “What out of-of-cycle write-downs may mean for real estate yields.” MSCI Blog, April 3, 2020.