Why liquidating a portfolio isn’t a zero-sum game

How has regulation changed the way portfolio managers and risk managers need to think about liquidity monitoring? 

Hassan: For a long time, regulators were focused only on how long it would take to liquidate a certain percentage of a given portfolio. This was usually calculated by aggregating position level liquidity metrics such as days-to-trade, which measures how many days it would take to trade out of a position and calculating an overall liquidity score. The problem was that it didn’t consider the other impacts of liquidating a portfolio. Now, the question being asked by regulators is: How long would it take to face a redemption, representing a percentage of the portfolio and what would be the cost? This is a far more multi-dimensional question compared to the original ask. 

What are those other impacts and dimensions?   

Joseph: There are at least three: the liquidation horizon, which is the original consideration; the cost of the liquidation; and the impact on the risk profile, which is the one that is often overlooked. At the end of the day, the question shouldn’t really be about whether you can liquidate a portfolio or not, because in most cases, you can, at a given cost. The challenge is liquidating while minimizing the cost and within your level of tolerance for risk profile distortion. It’s all about the trade-off. 

So, then how do you manage this give and take?  

Hassan: Ultimately, you need to map out some kind of a liquidity frontier. In a recent whitepaper, 'How Long Will It Take to Liquidate a Portfolio – And What Will It Cost You?’ we demonstrated how to do this, step-by-step, using the Axioma Portfolio Optimizer, a very flexible problem solver. We ran a variety of optimizations across different scenarios for both an equities and fixed income portfolio, so you can easily visualize the different interactions. 

Can you give a real-world example that you could optimize?

Joseph: Sure, let’s take the ESMA 5/10/40 rule for example. This is a diversification requirement for all UCITS products and is intended to mitigate issuer concentration risk. Under the rule, UCITS funds cannot invest more than 10% into any issuer and the total weight of all issuers in which it invests into 5% or more, cannot exceed 40%. What ESMA wants to know is the maximum amount that could be liquidated within one week while still respecting the 5/10/40 rule. The Axioma Portfolio Optimizer can give you that answer (and a range of other options). 

How would you approach this?  

Hassan: ESMA proposes running a reverse liquidity stress test. But with our tools, you can go one step further. We first used the days-trade-metric set at a ≤ 5 liquidation period which we extracted from our enterprise risk system, Axioma Risk. Then we used the Axioma Portfolio Optimizer’s integrated 5/10/40 module to divide the portfolio assets into two separate sets: A liquid one and a less liquid set. In our paper, we ran the illiquid set through four market scenarios: Normal Market Conditions, OAS + 50bps, Bid-Ask Spread x 2 and Trade Volume/2 scenarios.  

In the whitepaper, you write about horizontal and vertical slicing. Can you tell us more about this?  

Joseph: When you try to partially liquidate a portfolio, there are typically two competing objectives. The first one is to keep the portfolio’s risk profile as unchanged as possible. And the best way to do this is to just trade all the assets proportionally. That is what’s called vertical slicing. The second objective is to reduce the market impact. Intuitively, the most illiquid positions will have the most market impact, so you trade the most liquid positions first. That is known as horizontal slicing. However, doing this would then have knock-on effects on your first objective, which is about minimizing portfolio distortion and vice versa.  

What if you wanted to understand the market impact of liquidation ratios and trading horizon?  

Hassan: It’s all possible. The framework presented in the paper allows you to quickly meet any of the liquidity regulations but it can also help you assess if your portfolio is liquid enough, the suggested trades and the impact of the risk profile. You can also add as many constraints as you want into the Axioma Portfolio Optimizer, for example, maximum tracking error, the number of securities and maximum security weights. The point is that a liquidation exercise is not a zero-sum game. By using the right tools, you can meet your overall objective without having to sacrifice tracking error for cost or the other way around.  

For more details on how the Axioma Portfolio Optimizer can help your liquidity needs, download the whitepaper, How Long Will It Take to Liquidate a Portfolio – And What Will It Cost You? or contact us.