The Case for Cash

This article was featured in CFA Magazine's July-August 2012 edition.

In 1970, Warren Buffett opened a family safety deposit box to discover a letter from his grandfather and $1,000 cash. The letter, written to Buffett’s uncle on his 10th wedding anniversary, extolled the virtues of ready cash, explaining, “Over a period of a good many years I have known a great many people who at some time or another have suffered in various ways simply because they did not have ready cash. I have known people who have had to sacrifice some of their holdings in order to have money that was necessary at that time. Thus I feel that everyone should have a reserve.” He continued, “It is my wish that you place this envelope in your safety deposit box, and keep it for the purpose that it was created for.”

As Ernest Buffett explains, managing strategic liquid reserves is an important component of risk management. A cash reserve should be considered as an asset class that provides very distinct advantages to client portfolios. Three in particular are paramount: risk management (a tool to manage downside capture and mitigate volatility in markets where valuations are unattractive), opportunity management (dry powder that can be efficiently deployed to take advantage of investment prospects offering superior risk-adjusted returns), and liquidity management (hedging known distribution requirements against unknown market forces).

Given these advantages and Ernest Buffett’s more commonplace example, liquidity management makes sense conceptually. By immunizing near-term spending requirements in a proper way, clients are empowered to embrace a long-term investment approach. However, cash and ultra-short duration investments are typically overlooked in portfolio management—particularly in an environment where yields are flat and potentially negative when adjusted for inflation—because of many investors’ tendency to ignore short- and intermediate-term risks, the proclivity to emphasize upside capture over downside protection, and false expectations that a “total return” investing strategy will always provide for distribution requirements.

Managing the amount of cash reserved for distributions is separate and distinct from managing volatility across the broader portfolio. While some amount of cash (or liquidity) will dampen overall volatility, cash reserves are not intended to protect against downside capture or tail risk to any great degree. Cash held for distribution is a strategic investment decision based on distribution requirements and market volatility.

Liquidity management can be broken down into two distinct categories—internal and external liquidity. An internal liquidity requirement refers to the reallocation of assets within the portfolio and includes capital calls (private capital) and the deployment of cash across primary investment building blocks (safe assets and riskier assets) as part of a staged investment plan (for risk management or opportunistic reasons). External liquidity needs (spending/ distribution requirements) are monthly or periodic distributions from the portfolio that would draw down the overall principal balance, including tax payments, debt servicing, required distributions from IRAs, mandated foundation spending, obligations for endowments, and so on.

In both internal and external instances, the goal of the cash reserve is to eliminate the uncertainty around the source of the funds over the short- to intermediate-term, thereby “immunizing” distribution requirements. Although it is true that holding cash during a rising market dampens potential positive returns, this is merely an opportunity cost, not a realized loss. But liquidation of assets during a falling market can exacerbate negative returns and create a realized investment loss that directly impacts the portfolio’s rate of return.

For example, if a client distributes 2.5% every six months from the portfolio (and does not maintain a cash reserve) he or she will be required to sell assets in order to generate the cash needed for distribution. If this sale were to take place during a severe market downturn (say, a 10% decline), the portfolio would need to generate an extra 0.25% in investment return to compensate for the loss on the distributed capital. When combined with investors’ predisposition to loss avoidance (some studies suggest that investors prefer avoiding losses twice as much as experiencing gains), this hurdle is a powerful argument for the need to immunize distribution requirements from uncertainty and thereby create a cash reserve portfolio.

After making the decision to hold cash, the next step is to determine how much cash to hold in reserve. The method used at my firm, Balentine, is to estimate the amount of money the client needs for both internal and external purposes over a 12-month period. We also want to determine whether that amount will remain constant or be adjusted for inflation over time. Once both the amount and rate are
established, the next step is to formulate a policy dictating how much to allocate to cash and/or safe investments.

Another key consideration is the extent to which the cash needed for distribution requirements can be generated organically within the portfolio through either dividends or income. A critical (and often overlooked) piece of the puzzle is the sustainability and diversification of the yield as well as the frequency of the yield contribution. While the total annual yield may sufficiently cover the distribution requirements, the timing of the inflows may not align with the desired distribution schedule. Therefore, even if the client’s bond portfolio pays exactly the amount of an annual distribution, the trade-off will not always be equal. If the client withdraws a certain amount every month but receives only the interest to cover these costs every six months, an imbalance will exist for at least six months. Thus, depending solely on the interest income for the spending needs would not be practical.

A sound policy is to always maintain one year’s worth of required cash in reserve. Our global-macro risk outlook typically dictates the level held in reserve beyond one year. Obviously, a higher-risk environment warrants a higher level of allocation to liquid and safe investments.

Proper risk management dictates that 12 months of spending should always be kept in the cash reserve. Eliminating the uncertainty around future distribution requirements significantly outweighs the potential opportunity costs of holding assets with principal risk (e.g., low credit quality and/or higher volatility). But because cash can act as a “drag” on performance, the best approach is using safe asset classes (e.g., nominal and inflation-linked fixed income) as a container for any forecasted distribution requirements beyond one year. The purpose is to keep portfolios as nearly fully invested as possible without exposing future spending requirements to low-quality or higher-volatility asset classes, which would potentially put the clients who receive distributed capital at risk.

Of course, safe assets (those that are of higher quality and maintain a short to intermediate duration) are not without their own risks. But the tendency for bonds to experience hypervolatility pales in comparison to that of traditional market assets. Safe assets are a reasonable compromise between risk and return.

The need to hold cash is clear, but it is also important to develop a plan for refunding a reserve. The frequency and the source of replenishment for this reserve ultimately dictate when and how the cash should be refunded. The source of the funds will determine when to refund. In a neutral risk environment, by holding the second year of spending needs in safe assets, the probability of having to
refund the cash reserve from a severely depreciated asset class is greatly reduced. If the situation were reversed and safe assets were underperforming, then riskier assets should be the source of funds. At worst, this situation creates a choice between two uncorrelated asset classes and requires “forced rebalancing,” which affords the opportunity to harvest volatility by selling high and buying low. By refunding the cash reserve from the better-performing asset class, the portfolio is selling appreciated assets rather than underperforming assets. This result provides an additional level of risk management.

The question of when to replenish the cash reserve is more arbitrary and is tied to environmental risk outlooks. Typically, a good policy is to replenish the cash portion of the reserve when the level drops to three months of spending. If the process is begun with three months of liquidity in the cash reserve, the portfolio should have time to ride out any temporary market events (e.g., liquidity crisis). While exact timing is difficult, the frequency of refunding should correlate directly with the risk outlook. For example, as environmental risks increase, the frequency of refunding should increase. As global risks rise, the probability for asset prices to erode increases as well.

As managers consider implementation, it is important to remember that not all cash is created equal. In addition to money held in checking or savings accounts, marketable fixed-income instruments with a maturity of less than one year are generally considered “cash.” Limiting cash to investment-grade securities with a short duration, less than one year, ensures both marketability/liquidity and low risk of principal loss. Inflation risk is one of the risks of holding cash. This is particularly apparent in an environment where inflation exceeds short-term interest rates. Managing both liquidity risk and principal risk amplifies the potential for inflation risk, but short-duration TIPS could be used to manage the inflation risk in certain environments.

Low principal risk does not necessarily imply low liquidity risk. For example, a bank CD has very low principal risk but high liquidity risk. But low liquidity risk coupled with low credit risk and low duration risk usually implies low principal risk. Therefore, managers should be willing to slightly relax the duration risk constraint in an effort to preserve purchasing power in certain environments (extraordinarily low interest rates with normal or high inflation), provided that credit risk is kept to a minimum. This exception allows for an extension of duration beyond the three months typically used by money market funds. Although allowing duration to extend up to 24 months in a cash-type vehicle can somewhat reduce the opportunity cost of holding cash, a rapid rise in interest rates could negatively impact the value of the longer duration cash fund.

Because various risks can be opposing forces with one another, it is not possible to manage all the risks present in a cash portfolio (inflation, principal, interest rate, credit, liquidity, and duration) simultaneously. The primary purpose of a cash management policy is to prevent a shortfall in immediate- and near-term spending needs. Thus, a marginal amount of duration risk provides an opportunity to mitigate most of the risks without much cost. No matter what the situation, an inherent amount of risk is a given. The strategies I have outlined should alleviate some liquidity-management worries and provide principal protection, even during the most volatile times. Immunizing near-term spending requirements in a proper way empowers investors to embrace a long-term approach to portfolio management.