Commitment Approach vs. VaR (UCITS Leverage Calculation)

To calculate the exposure of a UCITS fund, the commitment approach estimates the leverage caused by all the derivatives in the portfolio. In contrast, the VaR approach measures the maximum potential loss within a specified time period due to market risk, rather than leverage.

That’s the gist of it.

For a more comprehensive explanation, keep reading:

What Is UCITS

UCITS (Undertakings for Collective Investment in Transferable Securities) is a set of directives that define the rules and regulations for investment funds in Europe.

An investment fund gathers investors’ capital and invests that capital in a portfolio of financial securities.

In essence, a UCITS fund is a collective investment fund easily accessible by retail investors. Why?

Because with this directive, the European Commission aims to give funds a European passport. Meaning, if a fund is approved in a single member state, it can operate and be marketed to retail clients all over Europe with no further authorization needed.

It promotes the EU’s goal of a single market for financial services in Europe, as all UCITS funds are subject to the same regulation in each member state.

UCITS funds are generally considered safe for their 5/10/40 diversification rule:

• Can’t invest more than 5% in securities from the same entity.
• Those 5% can reach 10% if that total investment in that entity doesn’t surpass 40% of all assets of the UCITS fund.

Additionally, UCITS funds invest mostly in transferable securities—liquid securities traded on capital markets (therefore, easy to price) that have adequate public information available.

Directive 2009/65/EC legislates UCITS. So any time you see it mentioned in the prospectus of a fund, it is a generally UCITS fund.

Like all European directives, this is not a law per se.

Each country does as it pleases when it comes to applying the directive into local law. For example, in Luxembourg, Part I of the Act of 17 December 2010 governs UCITS (meaning, it converts directive 2009/65/EC into local law), while in Spain it is law 35/2003, and in Germany, UCITS are referred to as OGAW (Organismen für gemeinsame Anlagen in Wertpapieren), while in France it’s OPCVM (Organisme de Placement Collectif en Valeurs Mobilières).

All funds domiciled in European countries and not following the UCITS directive are Alternative Investment Funds (AIF), under the AIFM directive.

To limit risk, UCITS funds must calculate their global exposure every single day.

Global exposure refers to the market risk of the fund’s portfolio, as well as the leverage generated by derivatives it trades or securities lending activities.

UCITS funds have to choose between the commitment approach or the VaR approach to monitor exposure:

UCITS Commitment Approach

Leverage is any method that increases the fund’s exposure. It’s when the fund borrows cash or uses derivatives to amplify its returns.

Under UCITS, the leverage calculation includes derivative positions and securities financing transactions only.

This means there is no regard for market risk. A fully invested portfolio without derivatives has a leverage of 0%. Similarly, netting and hedging may reduce leverage to zero (more on this below).

Calculating leverage under the commitment approach starts with converting all the derivatives positions in the portfolio to the market value of an equivalent position in the underlying asset.

This conversion varies according to the type of derivative. But it usually is the notional contract size times the number of contracts.

The next step in the commitment approach calculation is to consider:

• Hedging arrangements: Derivative instruments used for hedging purposes are subtracted when calculating the global exposure of the UCITS. To be considered hedging arrangements, these instruments cannot add any incremental market risks.
• Netting positions: Opposing transactions that refer to the same underlying asset are offset against each other.

This gives you net leverage, which is the goal of the commitment approach.

In contrast, gross leverage is simply the sum of the notional of all positions. It is a more conservative way to measure exposure. Here’s why:

The notional refers to the total value of the underlying asset in the derivatives contract.

Depending on the type of derivative, it can be the total value of the position, how much value the position controls, or an agreed-upon amount in the contract.

Gross leverage is based on the sum of notional exposures of derivatives—including those held for risk management. There’s no distinction between derivatives held for speculation and those for hedging (like when calculating net leverage). The result?

Unlike in the commitment approach, even strategies that aim to reduce risk are considered to increase the leverage level of the fund.

UCITS funds using the commitment approach calculate net leverage every day.

These funds must ensure global net exposure relating to derivative instruments does not exceed the total net asset value (NAV) of its portfolio. In other words, UCITS cannot be leveraged in excess of 100% of the fund’s NAV. These funds are thus unleveraged.

The NAV is simply the difference between the assets and liabilities of a fund. You calculate it by adding up what a fund owns and subtracting what it owes.

See also: Do You Understand the Difference between NAV and AuM?

Now let’s move on to the VaR approach:

UCITS VaR Approach

When UCITS funds engage in complex investment strategies (use a lot of derivatives in their portfolio), they often opt for the Value-at-Risk (VaR) approach. It is a more advanced risk measurement methodology to calculate global exposure.

VaR tells you the potential loss for a portfolio and the probability that that defined loss will occur.

There are three key components to calculating VaR: Time frame, confidence level, loss amount.

Essentially, it answers the question What is the most the portfolio can lose over the next day, month, or year? assuming a 95% or 99% confidence level.

For instance, a fund may have a 5% chance (95% confidence level) of the portfolio declining in value by 10% with a one-month time frame.

To estimate the VaR, the three most common models are:

• Historical: Look at prior returns history and order them from worst losses to greatest gains. The main idea is that past returns hint at future performance.
• Variance-Covariance: Also called the parametric model, it assumes gains and losses follow a normal distribution.
• Monte Carlo (most most common): Projects future returns by simulating thousands of scenarios using computational models stressing different parameters that impact the value of the portfolio.

Each model has its own set of assumptions, some being better depending on the fund’s portfolio. For instance, a Monte Carlo simulation model is better for a UCITS investing in derivatives with a non-linear risk profile than the parametric model.

Whichever VaR model the asset manager chooses, it should capture all market risks (interest rate volatility, sector-specific risk for equities, FX, or spread risk for example) that have a significant impact on the portfolio’s value, including those associated with financial derivatives (for instance, option greeks, or term structure risk).

When it comes to using the VaR approach to manage the exposure of a UCITS fund, there are two types:

Relative VaR Approach

Here VaR is calculated and compared to a benchmark.

The relative approach establishes a ratio of up to 200% between the VaR of the portfolio and the VaR of a reference portfolio (a market benchmark, such as the MSCI World Index for example). In other words, VaR of the funds’ portfolio cannot be 2x bigger than that of the benchmark.

This approach does not measure the leverage of the portfolio.

Instead, it allows UCITS to double the risk of loss of a reference portfolio. The similarity of risks between both portfolios prevents the UCITS from using highly leveraged strategies.

This is because of the following requirements when choosing the reference portfolio:

• Must be unleveraged.
• Must have a similar risk profile to the UCITS’ portfolio.

Thus, the benchmark is always a similar portfolio with no derivatives, such as an index or a fictitious comparable portfolio.

The UCITS directive requires that the VaR calculation is based on a time horizon of one month and the confidence level is 99%. Also, all positions and derivatives are considered, including derivatives with the goal of mitigating risk.

Absolute VaR Approach

When there is no benchmark, funds typically use the absolute VaR approach.

Here, the fund sets a maximum acceptable VaR for its portfolio relative to its NAV.

The higher the VaR it chooses, the riskier the fund becomes. Why? Because it has more room to allocate money to volatile securities, which results in a higher VaR number as these securities lead to higher potential losses.

For UCITS using this approach, the VaR of the portfolio cannot be greater than 20% of the fund’s NAV (calculated at a confidence level of 99%, with a time frame for the potential maximum loss of 1 month, and historical data going back at least 1 year to feed the forecasting model).

In other words, the maximum potential loss within a month with a 1% chance of occurring cannot exceed 20% of the difference between the fund’s assets and obligations.

What will determine this though? How exactly is portfolio VaR calculated?

Usually with a Monte Carlo simulation.

It simulates thousands of future scenarios for different risk factors and the relationship between them. Based on what? Historical data (in this case going back at least a year). Then, it estimates the behavior of the portfolio as a whole in response to each possible future scenario.

In essence, it predicts future investment prices through a model and then uses statistics to determine the worst-case loss on the portfolio.

Management companies always have to disclose the method they use to calculate the fund’s global exposure in its prospectus.

Having gone through in detail each approach to estimating global exposure, we can now drill down on the difference between VaR and commitment approach:

UCITS Commitment Approach vs. VaR

The commitment approach measures the risk exposure of investing in derivatives (taking into account the netting and hedging effects).

Each derivative is converted to the market value of an equivalent investment in its underlying asset. Depending on the type of derivative, there may be complex rules to make this translation.

Sum all that and the total equivalent investment in underlying assets cannot exceed the funds’ NAV.

Therefore, the commitment approach focuses on limiting the leverage the funds use.

In contrast, the VaR approach does not look at leverage.

Instead, it looks at the risk of the portfolio (which includes stocks, bonds, and other securities, but also derivatives) of the fund losing value. How so? By running thousands of simulations of how various risk factors would affect the portfolio.

Thus, the VaR approach focuses on limiting the market risk of the portfolio seeing a drop in value due to unfavorable market fluctuations. There are two ways to do this:

• Relative VaR: The VaR on the UCITS portfolio cannot exceed twice the VaR on a comparable benchmark portfolio.
• Absolute VaR: The VaR of the UCITS fund is capped as a percentage of its NAV. It is not compared to a given benchmark.

But how do funds choose between commitment, relative VaR, or absolute VaR approach?

Based on the investment strategy of the fund and on the type, complexity, and proportion of the derivatives it trades.

In essence, a UCITS fund using complex investment strategies (for instance options strategies, arbitrage, long/short) or with significant exposure to derivatives should use the VaR approach to calculate global exposure. Why?

Because the commitment approach does not adequately capture risks related to derivatives, such as volatility risk, gamma risk, or basis risk.

On the flip side, if the risk/return profile of a UCITS changes frequently or if it’s not possible to define a reference portfolio, the relative VaR method should not be used.

A UCITS that chose to use absolute VaR cannot switch to relative VaR (or vice-versa) simply because it breached the limits it set.

Regardless of the method chosen, to comply with the regulation, the fund must establish, implement, and maintain a documented system of internal exposure limits.

Let’s go through a VaR vs. commitment approach example to put these concepts in motion:

UCITS Leverage Calculation Example

Let’s say you work for a big asset manager managing a €100M NAV UCITS fund domiciled in Europe.

The fund aims to replicate the performance of the MSCI Emerging Markets Asia Index by directly investing in its constituents (large and mid-cap stocks from Asian emerging markets).

Little side note: Investing in emerging markets is riskier. Why? Because there is a higher chance of political or social instability, less publicly available information about financial instruments, and less supervision on accounting and financial reporting.

Since the fund is domiciled in Europe but invests in Asia, there’s also currency risk.

This is because the value of the fund’s investments will not only fluctuate according to the market, but also according to exchange rate changes.

For this reason, the fund will use Currency Swaps (FX) to hedge this risk.

More specifically, it will pay a fixed amount of EUR every month on a €40M principal amount and receive a floating amount of CNY (Chinese yuan) on ¥315M.

The fund and its counterparty (usually a big financial institution) will not actually exchange the principal. It only serves as the notional value upon which the interest payments to be exchanged are calculated.

What happens is when the Yuan loses value against the Euro, the fund’s portfolio loses value as well. However, the loss is reimbursed because the asset manager receives interest payments at whatever rate the Yuan trades at the time, thus benefiting from its increase in value.

And if the Yuan increases in value against the Euro, the portfolio gains value even if the price of the stocks stays the same.

The thing is the asset manager agreed to pay a fixed Euro interest rate, so it is “stuck” paying a higher interest rate, canceling out the increase in value the portfolio had thanks to the rise of the Yuan. The result?

Currency fluctuations are no longer a factor for portfolio value.

The rate is fixed at 1/40/315=7.875EUR/CNY from the moment the swap is put in place.

The fund will also use Total Return Swaps (TRS) on commodities. These provide exposure to the performance of commodities markets without investing directly in physical commodities.

The underlying market value of the reference asset (commodities) is €30M.

How do you compute the global exposure using the commitment approach?

• Commitment: Calculate the exposure of each individual derivatives contract. That is, the equivalent position in the underlying asset or the notional value (the latter is more conservative).
• Netting or hedging arrangement: Subtract the market value of securities relating to opposing positions in the same derivative, or used with the goal of hedging.

For the FX swap, the notional value of the leg the fund has to pay is €40M. As to the TRS, we know the equivalent position is €30M.

Thus, the global exposure is 40+30=€70M. But since we offset derivatives for hedging purposes (in this case the currency swap), the global exposure under the commitment approach is €30M.

This value is well below the NAV of €100M, meaning the fund is complying with the UCITS rule.

Now, how do you compute the global exposure using the VaR approach?

Remember, the VaR approach is a measure of the maximum potential loss due to market risk rather than a measure of leverage.

You start by computing the VaR of the fund’s portfolio by running a simulation stressing a number of risk factors. This is the Monte Carlo simulation model.

The goal is to find the maximum potential loss, including all derivatives positions regardless of their purpose.

Starting with the relative VaR, you compare that value to the VaR of a benchmark. If it is not double? You’re good.

As to the absolute VaR approach, the UCITS’ portfolio VaR needs to be below a predetermined limit (at most 20% of NAV to comply with UCITS regulation).