Analysis

Private capital’s technology tipping point


technology data reflected on spectacles scaled

Corporate history is littered with examples of companies that dominated their industries for decades but neglected embrace technology and innovation and paid the price.

There are some very well-known examples: The now defunct video giant Blockbuster passed on Netflix, the once unprofitable start-up DVD postal-renting service with a current market capitalization of close to US$270 billion. Netflix disrupted its own model and utilized internet technology to produce on-demand content.

Kodak, who filed for bankruptcy in 2012, developed the first digital camera back in 1975. It was reluctant to reposition the business towards digital given the high amount of investment required and was swept aside amid the development of smart phones, tablets, and the global embrace of digital photography.

Regarding smart phones, Nokia sowed the seeds of its own decline by not recognizing that app-based software would form the basis of the mobile phone’s future. There is clear causal relationship between its focus on physical devices and its fall of market share from 51% to 5%.

Slightly closer to home for private markets is the ongoing battle for digital supremacy in the global banking sphere. The rise of so-called ‘challenger-banks’, who offer streamlined, digital-first retail banking services, has undoubtedly disrupted the industry.

The vast majority of incumbent, traditional banks – both global and regional – were incredibly slow to adapt to digital models for their consumer servicers. With legacy IT applications prevailing and technical debt building up, banks lack of clear-sighted strategy to deal with both digital transformation and changing user needs and expectation was evident at the tail end of the last decade.

It is of course incredibly unlikely that digital players such as Atom Bank, Tandem, Monzo, and Starling Bank, Revolut will replace traditional banking players. The size of wallet-share and financial resources incumbent players have is vast compared to that of the challenger banks.

However, what it has done is rapidly accelerate the digitization of these institutions in efforts to ensure account balances, reduced cost-to-income ratios, higher customer acquisition and retention either stabilize or grow. From transforming their operational technology infrastructure to building customer-centric apps, banks have had to follow where more-nimble and agile ‘new-banks’ have trod.

Indeed, challenger banks development of seamless user experiences, quick and easy account registrations have all been aped in the broader market. No banking app would be trusted without the enhanced security measures pioneered by digital first entities, and the integration with third party applications is a trend that’s set to continue across all modern platforms.

As digital services take an ever firmer grip over financial institutions’ product suites, how customer data is managed and how customer needs are need met with that technology will undoubtedly determine which banks take a greater share of future customer dollars.

What can private capital learn?

Private capital is now facing a similar technology tipping point which has the potential to reshape the long-term make-up of the market. Managers that take on the opportunities provided by cloud computing, data analytics, automation and AI will thrive. Managers that wait too long will fall by the wayside.

There are multiple areas where technology is transforming how private markets managers run their businesses:

1. Data management and investor reporting

Global private capital assets under management (AUM) climbed to US$14.5 trillion in 2023, according to Bain & Co, more than triple the levels from a decade earlier in 2013.

The rapid growth in AUM has meant more funds, more transactions and higher reporting expectations from investors who now have a much bigger exposure to alternatives to manage. With more capital at work in private markets, investors are demanding more bespoke, granular data and more frequent reporting on portfolio performance.

This demands that managers upgrade their reporting infrastructure, and technology is a key enabler of this.

Private equity’s roots lie in small, nimble teams of dealmakers with low overheads and small back-office teams. As the industry has grown, however, managers have had to invest more in back-office support to keep up with investor expectations.

Firms that move early to harness technology, outsource or co-source back-office functions, invest in cloud-enabled infrastructure overlaid with best-in-breed fund accounting technology will pull ahead of their peers. Factor in utilizing data warehousing and taking advantage of advanced data analytics and the gap will widen further still between those who have and haven’t transformed their business. 

2. Rising regulation

Growth in AUM has also led to a rise in regulatory scrutiny and step-change in regulatory reporting, compliance and disclosure.

Alongside higher expectations around investor reporting, closer regulation has placed additional pressure on manager operating and finance models.

A failure to take advantage of the technology available in the market to drive back-office efficiency and keep compliance costs in control will lead to managers having to spend more senior resource and time on regulation and compliance and less in the core business of sourcing, managing and exiting assets for the best returns.

3. Tech-enablement of the front office

AI, automation and predicative analytics are also transforming how front office functions work, with firms utilizing these tools to free accelerate deal origination and free up dealmakers to spend more time on the high-value tasks of building relationships with deal targets and running negotiations.

First-mover in the industry already have AI-enabled platforms in place that allow deal teams to expedite deal selection, benchmark valuations, monitor sector trends and combine third-party and inhouse data into a single data repository that can be mined to assess deal opportunities and get dealmakers up to speed on new deals at pace.

Managers harnessing these tools report significant benefits, with some claiming that technology has helped them to identify deal targets as much as a year before peers and achieve superior deal conversion ratios.

4. New fund structures and a broader array of investors

Private markets have a growing history of innovation where it comes to developing new ways of organizing and attracting investment. One such example is the recent momentum behind ‘Open ended’ funds (OEF), which have given retail investors an access point to alternative markets.

While this ‘democratization’ is a positive, standing up and managing an OEF is rife with complexities: from more investors and more redemption requests to a huge increase in producing accurate NAV calculations, the management of these funds can be challenging. The right technology is key to the success of these vehicles.

At the other end of the wealth spectrum, the appetite for High Net Worth Individuals (HNWI) to dip their toes into alternative waters has also grown, no doubt spurred on by the above average returns of private markets investment and an opportunity to diversify their portfolios.

Indeed, Boston Consulting Group estimate that by 2025, HNWI in private equity alone will rise to a staggering $1.2 trillion. Each HNWI’s needs may differ wildly in terms of data requirements and reporting. Meeting their needs will not be done effectively with legacy tools and systems.

The right partner, armed with the right technology

Managers do not have to undertake this transformation of their business structures alone, tech-enabled service providers such as Alter Domus have the tools and experience to support managers through this technological inflection point.

As trusted partner to hundreds of managers, Alter Domus has developed deep, lived experience of what is required to upgrade technology and operational infrastructure in practice.

The administration and fund accounting infrastructure that would have been perfectly adequate to manage a private equity fund 15-years ago is now no longer fit for purpose. Best-in-class technology and digitally powered fund operations have become essential for private market stakeholders.

Alter Domus’ Digital Workflows Application was developed as a response to the market need for a transformative technology. Workflows is designed to handle the volume and complexity of private equity funds and builds a “digital bridge” between client and fund administrator. Clients are already reaping the benefits of this market leading capability; as our partners at leading asset management house, Coller, have commented:

Alter Domus Digital Workflows application has significantly improved our fund administration experience. Its utilization of automation and AI, combined with digital access to each part of the process is enhancing our transparency, operational efficiency, and data accuracy

Coller Capital

As well as the enhanced transparency across our clients’ fund portfolio, Workflows ends the reliance on outdated communication channels such as email and phone calls and turns data into an analysis-ready single source of truth.

At Alter Domus we understand that the investment required to keep pace with technology and innovation, coupled with the risk of disruption to process that have underpinned success for decades can be daunting. However, the consequences of not acting are far more severe. Alter Domus is here to aid and support our clients in taking those next, transformative digital steps.

Key contacts

Demetry Zilberg

United States

Chief Technology Officer

Insights

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

architecture glass tress
EventsApril 22-24, 2024

BAI Alternative Investor Conference

Analysis

Leadership means business: why gender diversity is a commercial imperative

There are strong ethical reasons for creating gender diverse workplaces. The commercial argument for gender balance is equally compelling. 


people informal at meeting

When it comes to gender diversity in the workplace the private equity industry hasn’t always been at the vanguard of reform. The good news is that things are changing.

According to The state of diversity in global private markets: 2023, a landmark annual review of diversity in private markets compiled by McKinsey, just under half of entry level roles in private equity were held by women at the end of 2022.

A separate report from the British Private Equity & Venture Capital Association (BVCA) and Level 20 (a not-for-profit organization campaigning for gender diversity in private equity) found that women held 11 percent of senior investment roles in private equity in UK firms and their European offices – up from 10 percent in 2021 and just six percent in 2018.

In an industry where there has been limited female representation historically, these findings are signs of encouraging advancement to a more gender-balanced private equity work force.

A commercial priority

Improvements to the gender mix in private markets, as well as the wider corporate world, reflect an acknowledgment by businesses of the ethical argument that workplaces should reflect the demographics of wider society. Alongside the ethical case for gender diversity in business settings, however, there is also a growing bank of evidence showing that gender balance isn’t only a moral priority but commercial one.

Zenger Folkman research referenced in Harvard Business Review, for example, found that women rated better on key leadership capabilities than men. A separate study by S&P Global Market Intelligence showed that companies with female chief financial officers outperformed their sector averages, and that businesses with female chief executives generated superior share price performance to peers.

Academics have also found that organizations with female leaders in the C-suite are more likely to prioritize investment in research and development and are more careful when taking risk. There are several other similar studies highlighting the commercial benefits women bring to corporate leadership.

Diverse teams drive private markets performance

The evidence pointing to superior corporate performance when senior management is more diverse is mirrored in the private markets industry. Research led by HEC professor Oliver Gottschalg, utilizing a data set of close to 2,500 deals executed by 51 managers over 20 years, found that buyout teams with at least one female member outperformed all-male teams on IRR and total value to paid in (TVPI) metrics.

Investment committees with at least one female representative, meanwhile, were also shown to deliver better outcomes than male-only committees, with IRRs higher by an average of 12 percent and cash returns better by 52 cents per dollar invested for mixed teams. Gender balanced private equity teams were also shown to reduce risk, bringing down average capital loss ratios for funds by 8 percent.

“Our study is the first to prove empirically that performance of gender-balanced investment teams correlates with higher returns in private equity,” Gottschalg said in an interview on his findings.

Researchers put forward a variety of explanations for the findings of outperformance by gender-balanced teams, including a broader range of insight and perspective when assessing investment targets, improved decision-making and planning and a more balanced approach to risk.

Whatever the explanations for outperformance may be, it is becoming increasingly difficult to ignore headline research findings highlighting the added commercial value that diverse teams offer.

Untapped potential

But while private markets have made progress when it comes to opening up opportunity to female professionals, it is surprising that has not been faster when considering the growing body of evidence that teams with women leaders deliver outperformance.

Senior leadership in private equity leadership and investment teams is still male-dominated. McKinsey’s research, for example, found that while there was balanced female representation at entry level in private equity, at c-suite level women still only held 17 percent of senior roles. McKinsey models suggest that at the current rate of change it could take more than six decades before the industry achieves parity in investing roles at managing director level.

Almost 10 years on from its launch Level 20, meanwhile, is still only halfway towards its objective for women to hold at least 20 percent of senior positions in private equity.

It is important to note that it is only within the last ten years that private equity has recognized the value that recruiting more female professionals into their organizations can bring. Building up talent pools and attracting more women into the industry, however, is a long-term project. It takes time to develop candidates with the training and experience to take on deal making and leadership roles in what is an apprenticeship industry.

Level 20 and programs such as the 10,000 Interns Foundation are helping to address these bottlenecks, but with women still underrepresented in business schools, coupled with complexities around retaining women who have started families in the industry, there are long-term, secular challenges that continue to face the private markets industry when it comes to increasing female representation.

For private markets managers and service providers that have the conviction to tackle these obstacles, however, finding and promoting female talent presents a compelling commercial and strategic opportunity. For Alter Domus the value of gender balance in leadership is almost “old news”, and the firm and its clients have been benefitting from women in senior managerial positions for years, across all markets.

Sandra Legrand, for example, sits on the Alter Domus Group Executive Board and is Regional Executive for Europe and Asia Pacific, overseeing operations in 18 countries, including the Group’s headquarters in Luxembourg.

Under Legrand’s leadership, division revenues have grown by 123 percent, with assets under administration expanding to more than US$740bn, to rank Alter Domus as the number one fund administrator in Europe by funds closed.

Jessica Mead, Group Executive Board member and Regional Executive for North America, has delivered similar performance impact.

Operating out of Chicago in the US, Mead has overseen a transformative period of growth for Alter Domus, increasing headcount by more than 57 percent and growing assets under administration in North America to more than US$837 billion.

At operational level, meanwhile, Chief Human Resources Officer and Group Executive Board member Joanne Ferris has played a crucial role in supporting Alter Domus’ organic and inorganic growth.

Ferris has overseen the onboarding of more than 450 employees via M&A and other strategic initiatives, with retention rates for employees joining following M&A running at between 85 percent and 90 percent. Ferris has also managed 112 percent growth in global headcount to support client demand, as well as making provision for 269,000 hours of staff training and development.

As demonstrated through the leadership of Legrand, Mead and Ferris, gender balance and commercial imperatives are not mutually exclusive, but mutually reinforcing. Leadership means business and there is no such thing as ‘business’ without female leadership.

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

5 Trends Shaping Private Markets Secondaries in 2024

As the primary source of liquidity in an illiquid asset class, secondaries managers will have a crucial role to play as private markets emerge from a cycle of slowing distributions and fundraising.

Alter Domus highlights 5 key themes that will drive secondaries market activity through the course of 2024.


technology data on screen pencil in hand scaled

The last year has not been easy for private markets managers. Secondaries is one corner of the market that has bucked the trend.

In the midst of the myriad macro-economic headwinds that buffeted private markets in 2023, the secondaries space managed to deliver year-on-year growth across most key metrics, according to Jefferies figures.

Global secondary volume climbed four percent to US$112 billion in 2023, an impressive result considering the last double-digit declines in buyout and exit deal value over the same period.

Deal activity rallied particularly strongly in the second half of the year – with H2 2023 deal value 60 percent up on figures for H1 2023 –  as stabilizing interest rates helped improve LP portfolio pricing by an average of around four percent to 85 percent of net asset value, according to Jefferies. This helped to narrow bid-ask spreads and push more deals over the line.

The gathering momentum behind secondaries transactions through the second half of 2023 has carried in 2024, positioning secondaries strongly for the months ahead.

The outlook for M&A and IPO activity is brightening, but it will take time for these markets to get up to speed after a quiet 2023. LPs who have been holding out for distributions will continue to turn to secondary markets for liquidity, driving demand for secondaries investment. Against this positive backdrop, Alter Domus outlines five drivers of secondaries market activity in the coming months.

1. Resilient secondaries fundraising to drive pace of investment

Secondaries fundraising has proven remarkably resilient during the last 24 months, with Jefferies figures showing fundraising in 2023 exceeding the combined annual totals for 2021 and 2022.  The amount of capital now available for investment in secondaries is more than double the amount that has been deployed in deals in the previous 12 months.

The strong stock of secondaries dry powder means that managers have ample firepower to pursue deals. This will help to improve asset valuations, encourage more sellers to market and spur investment pace and secondaries deal volume.


2. GP-led volume to rally as more deals come to market

The primary driver of the growth in secondaries deal volume in 2023 was in the LP-led deal space, which climbed seven percent year-on-year, while GP-led volume stayed flat, according to Jefferies.

Moving into 2024, however, GP-led volume is positioned for an uptick. Managers remain pressed to make near-term distributions, and GP-led deals will be a key driver for doing that.

Through the course of 2022 and the second half of 2023 a number of GP-led deals launched but didn’t get done, as the delta between buyer and seller expectations in M&A markets spilled into the GP-led space.

These assets will still be prepped for GP-led processes. Investors will be able to see how assets have traded since GP-led deals were initially launched and valuations for sellers will have improved. These themes point to a strong uplift in GP-led deal flow.

3. The rise of multi-asset continuation funds

GP-led deal flow will also be spurred by rising volumes of multi-asset continuation funds.

Fund adviser Campbell Lutyens sees a surge in multi-asset continuation funds (in addition to ongoing activity in single-asset continuation funds) as they allow managers to realize liquidity from multiple assets in single transactions, securing distributions of significance for investors.

These deals have also gained traction with buyers, who have leant into opportunities to invest in carefully assembled portfolios of assets.

4. Managers and strategies will become more specialized

In order to differentiate in an increasingly competitive market, secondaries managers will become more and more specialized – by private markets assets and type of secondaries deal.

According to Campbell Lutyens, infrastructure secondaries deals accounted for 12 percent of LP-led deals by volume in 2023, nearly triple the levels observed in 2022. The market share held by private credit secondaries, meanwhile, doubled year-on-year to four percent. The growth of secondaries deals in private markets segments beyond the historic base in buyouts will see more managers specialize to gain competitive edge in specific segments of private markets.

Similarly, many managers will also narrow down their strategic lens to focus on either GP-led deals (where the focus is on selecting and diligencing specific companies in depth) and LP-deals (where the focus is constructing large, diversified portfolios of hundreds of assets).

5. Secondaries and GP-stakes convergence?

Early in 2024 Bloomberg reported that Blackstone’s GP stakes business (which invests in the management companies of private markets firms) would shift out of Blackstone’s hedge fund division and into the firm’s secondaries arm.

Secondaries and GP stakes strategies may seem quite different on first glance, but the two strategies do overlap. Private Equity International notes that both strategies involve a combination of due diligence on the manager of a target fund and the specific assets in that fund. Bringing the two strategies together can therefore unlock data and analysis synergies. The J-curves for both strategies are also similar.

There are potential conflicts of interest that can emerge in scenarios where secondaries and GP stakes strategies are combined, but these can be addressed.

The complementarity between two fast-growing private markets verticals could see more combinations of secondaries and GP stakes strategies.

Key contacts

Tim Toska

Tim Toska

United States

Global Sector Head, Private Equity

More events

technology man holding iPad showing data scaled
EventsApril 18, 2024

LSTA Operations Technology Conference

architecture round building
EventsApril 22-24, 2024

INREV Annual Conference

architecture glass tress
EventsApril 22-24, 2024

BAI Alternative Investor Conference

Analysis

Data, data, everywhere: how investors in alternatives can scale the information mountain


technology colleagues analyzing data on screen

The last decade has seen investment in private capital markets shift from a niche strategy to a core part of the fund playbook for asset owners of all types. Be it private equity, private debt, or real assets, the reasons why this investment shift has occurred are abundantly clear.

Firstly, investors can balance the overall quality and strength of their portfolio through diversification across these asset classes, which have broadly been providing above average returns. Secondly, the stock and bond markets have proved to be volatile hunting grounds, and alternative asset classes have proven to be a great hedge against this.

A long way up to the data peak

While they may differ in terms of scale and focus, wealth management arms of multi-national banks, multi-family focused managers, pension funds, and college endowments have all found success in the alternative arena. But as is often the case, success can be a double-edged sword and though these investors vary as institutions, they all face a common – and ever increasing – problem: data management.

As the phrase goes, for every action there’s an equal and opposite reaction, and as investors have increased the volume and size of funds within alt markets, the volume of data flowing in from asset managers has become mountainous. In conversations we here at Alter Domus have had with our clients, the story is the same: the data that comes their way is largely unstructured, in non-standard formats, and is generally not digitized.

This manifests itself in multiple problematic ways; not only is the operational task of aggregating and normalizing this data costly, slow, and labor intensive, but the manual nature of the process diminishes levels of accuracy.

No base camp for investors

A chief investment officer of a well-respected large multi-family investment house recently highlighted how the manual collation and fragmented delivery of data affects what happens in the front office on the front line of analysis.

“The challenge we face is that, as this piece-meal data flows in, we aren’t getting a clear picture of risk and opportunity across our entire portfolio. We need to trust our data. And that’s very hard to do that when there’s a chain of processes with far too many potential points of failure.”

These data problems are further compounded by everything from compliance and data security to regulatory demands. Factor in an over-reliance on legacy tools, from spreadsheets to emails to out-of-date accounting systems, and the multiple sources data flows in from, and the issue is deepened further still.

Ultimately, with allocations to alts rising, investors need to maintain their competitive edge. That ‘edge’ is the outcome of having a clear view of the operational performance of companies and assets within their portfolio, which in turn informs their current and future decision making.

Expensive Sherpas for the expedition

It would also be remiss not to talk about the labor issues that have developed because of the increase of capital flows into alternative markets. Outside of hiring in more investment professionals as funds scale, the operations, finance, and support teams that surround them to deal with data have exponentially grown too.

It’s been well documented that the labor market for talent in these areas has become ultra-competitive, meaning hiring and staffing costs have risen accordingly. Finding people with the requisite data skillset and understanding of markets is set to be an ongoing challenge.

Conquering the data Everest

So, what’s the solution to dealing with the data? The answer is two-fold; Firstly, businesses must employ and utilise a combination of market data knowledge and market leading technology. Secondly, companies need to look beyond their own walls to acquire that transformative combination. Put simply, building your own technology solutions, when your primary focus should be on your investment strategies, is prohibitively expensive, and time consuming to run, maintain, and constantly upgrade.

That’s why it’s imperative to partner with the right third parties who have the people and technology ready to go. And that’s also why Alter Domus has invested in building first class solutions that have been specifically designed to deal with asset owners’ data dilemmas.

The view from the top

Alter Domus Digitize – Investor Statements solution is designed to ingest any volume of complex investor documents, from capital calls to distribution notices and capital account statements, and to provide our clients with instantaneous access to actionable data. This managed service uses a combination of people, process, and automated technology to extract, validate, and deliver critical datasets to your preferred reporting systems downstream.

So how does it work?  Our clients’ fund managers distribute source documents to us via our workflow tool either directly or through their linked asset manager / GP portal. Our data team is notified to retrieve documents from the portal. Mailroom sends data to our automated extraction technology platform. Trained on hundreds of thousands of documents, it leverages machine learning (text and geographic based) to find the correct information in the documents. Data is pulled back into our workflow tool, saved, calibrated, and mapped to a list of fields which also standardizes and normalizes the data in the Alter Domus data platform. Both our automated system and Data operations team users perform double-blind checks and reconcile in case of any mismatches, ensuring both accuracy and quality of output.  An external transaction ID is created for audit trail. The workflow tool pushes data through to your reporting or other downstream systems.

Beyond the benefits in terms of costs, driving clear operational efficiencies, and changing our clients’ ability to scale their funds at will, receiving structured, digitized information at speed removes the impaired view of their portfolio that many investors have. With data now acting as a source of truth, from input to output, and across back to front offices, investors’ understanding of risk and performance across funds and assets is fundamentally transformed.  

Ultimately, we believe that investing in data excellence is the best way for investors to turn those data mountains into molehills. Find out how we can help you here.

Contacts

Alter Domus:
Brad Pratt
[email protected]


Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

Why stress testing and scenario analysis are vital in assessing pre-payment risk in broadly syndicated loans


Technology data on screen plus fountain pen and notepad

In the dynamic world of alternative investments, Broadly Syndicated Loans (‘BSL’) continue to be an important part of the market. The Leveraged Loans market as a whole yielded a near record 12% return this year. And CLOs are expected to continue to have high issuance in 2024, with most banks forecasting between 100 and 115 billion in new issuance this year, numbers quite similar to last year’s high level of issuance.

Amidst today’s high interest rates, it’s no surprise that the BSL market has become and continues to be a popular choice for investors seeking attractive yields and a relatively low risk profile. We’ve previously written about BSL characteristics, loss protecting features, and prepayments. We now additionally consider macroeconomic factors that drive prepayments. 

As the market grows in importance, investors are increasingly asking how BSLs might perform under a range of macroeconomic conditions. While some discussion has gone into monitoring default risk via stress tests, most stress testing is primarily focused on default risk and instead opts to zero out prepayments as a conservative assumption.

While this creates simplicity, it runs the risk of creating excessively conservative and unrealistic stress tests. This simplicity ignores an important fundamental concept in credit risk – the Exposure At Default (‘EAD’). EAD is a critical component of credit risk – in addition to probability of default (‘PD’) and loss-given-default (‘LGD’).

Using our Risk Modeler Analytics platform, we can adopt a more prudent approach to stress testing that considers both default risks and prepayment risks and ensures that EAD is addressed. Prepayment analysis is also generally critical for BSL investors for proper cash management, including investor redemptions and reinvestment strategies.

In our previous article, we discussed the benefits of forecasting prepayments and looked at a broad set of factors that can predict prepayment rates. However, we did not consider macroeconomic factors, which raises the question: Why might someone want to focus heavily on macroeconomic factors when forecasting prepayments?

Firstly, aggregate prepayments vary wildly across time, ranging from 3% to 12% on a quarterly basis in our sample portfolio. Secondly, these prepayments correlate strongly with macroeconomic factors in ways that are consistent with sound economic reasoning, suggesting that the correlations will hold in future scenarios.

Finally, this opens up the door to incorporating scenario-based thinking and stress testing as a means of driving analysis and decision-making. An important note to consider is that scenario analysis need not necessarily be only regarding stress scenarios.

Given the novel combination of interest rates and inflation in the current macroeconomic environment, numerous different positive scenarios can be constructed with different implications for prepayments. Considering such scenarios is increasingly important for sophisticated BSL managers attempting to do cash flow management.

Modeling approach

As part of our process, we selected a sample of loans from our BSL universe and attempted to model prepayments primarily using macroeconomic factors. We used two broad approaches to create our models.

The first approach, called the holistic approach, focuses on a wide range of macroeconomic factors. This approach allows users to potentially consider scenarios based on variations of any of these variables. For example, one might consider a high inflation/low-interest rate scenario, a high GDP/high inflation scenario, or any other combination. 

The second approach, called the targeted approach, focuses on a single specific variable along with its lags. This has the advantage of creating a parsimonious model, which is easier to understand and implement.

Holistic Model

The holistic model attempts to capture the macroeconomic factors that could affect loan prepayment speed. Inflationary pressure, macroeconomic outlook, and the interest rate environment are some of the macroeconomic factors that are generally expected to be relevant to BSL prepayments.

With this idea in mind, we assembled a large set of macroeconomic factors and tested them in combination with each other. Instead of testing all combinations of all transformations of the relevant variable pool, we tested combinations of the most important variables and avoided trying too many different transformations and combinations. The result is a model that includes several variables that are intuitive and logical.

The following variables are included in the model, along with an explanation of the economic reasoning for having them:

  • CPI (quarterly growth) –This variable captures the current quarter’s inflation rate. As such, it tells us about the current inflationary environment and provides a decent proxy for inflation expectations moving forward. Borrowers facing higher inflation expectations will be more inclined to find alternatives to prepayment (i.e. reinvesting in the business) and thus prepayments would be expected to decrease, all else equal.
  • S&P 500 (year-over-year growth) – As public equities markets are forward looking indicators of macroeconomic growth, the change in S&P valuations gives us a good indicator of how markets feel about macroeconomic prospects. When the S&P is improving, prepayments increase reflecting the fact that borrowers likely feel enough optimism to attempt to pay off more of their debts and/or raise equity financing to prepay debt.
  • High Yield Index Option-Adjusted Spread – The spread considered looks at yields of corporates with ratings of BB or less with respect to a spot Treasury curve. The index has a negative relationship with prepayments, all else equal. When spreads decrease, borrowers are less inclined to hold loans still tied to a higher spread from a previous high spread period. As such, they’re likely to prepay knowing that they have the option to get new financing under a more favorable spread.

In Exhibit 1 below, we see the model fit through historic data. In the earlier parts of the data, the sample of loans is not yet fully populated and consequently a bit more volatile from period to period. As such, the fit isn’t as strong. But from 2018 forwards, the model-fit is quite good and we can see that the model captures the major swings in prepayments that have occurred in response to the significant macroeconomic shifts that have occurred since 2018.

Exhibit 1: Macro Model Fit

Macro Model Fit - Quarterly Prepayment Rate

Targeted Model

While the holistic model captures a good deal of nuance regarding the various factors that could drive prepayment risk, the complexity creates some costs. Coefficients become harder to interpret as more variables enter the model.

Creating logically consistent alternative scenarios becomes harder as more variables must remain consistent with each other. As an alternative one could consider a simpler model that focuses on just one variable along with its lags. Fortunately, when one looks at the High Yield Option-Adjusted Spread, one finds such a variable.

Simply plotting the two variables against each shows a strong inverse relationship between spreads and prepayments. There’s also a straightforward and logical explanation. When spreads are lower, borrowers prefer to prepay and take advantage of lower spreads for refinancing. When spreads are higher, borrowers do not.

Exhibit 2: Spread vs Prepayment Rate

Spread vs Prepayment Rate - Quarterly Values

We can also consider lagged values of yield data, which points to an inverse relationship between spreads and prepayments. Lagged values are positively correlated with present prepayments. This can be explained by seeing current and previous spreads as both playing a role in driving prepayments.

If today’s spreads are low, but 6 months ago spreads were also low, then prepayments might not be as high as in an alternative situation where today’s spreads are low but 6 months ago spreads were extremely high. In the first case, borrowers likely took advantage of low spreads, but in the latter case, borrowers likely have high spread loans that they would be eager to refinance.

Exhibit 3: Target Model Fit

Quarterly Prepayment Rate (% of $)

Scenario Analysis

Creating forecasts across scenarios is necessary for doing scenario analysis. To do so, we start with a simple baseline forecast for High Yield Spreads, that shows it steadily decreasing over the near term. For alternative scenarios, we simply take the forecast and bump them up and down by 2 standard deviations of the series’ historic values. With that straightforward scenario prepared, we can create prepayment forecasts for our portfolio to see the impacts of changing spreads.

We can see the results below in Exhibit 4. Higher spreads lead to lower prepayments, lower spreads lead to an increase in prepayments. Interestingly, because we use previous values of spreads in the model, the model inherently recovers towards the baseline. This makes sense intuitively. If spreads drop, borrowers would prepay and refinance to take advantage of new lower spreads.

But if spreads remain at that level, borrowers will eventually no longer need to keep refinancing. By the converse, when spreads rise that may initially create difficulty for prepaying, but as borrowers adjust to the higher spread environment, they may find themselves reverting to their more typical prepayment behavior.

Exhibit 4: Scenario Forecast

Quarterly Prepayment Rate (% of $)

Out-of-sample Results

These models were developed in early 2023 with data only being available through 2022Q4. At the time of writing, both prepayment and macroeconomic data were available through 2023Q3, allowing us to evaluate the accuracy of our forecasts by comparing them to the actual data.

The results showed a notable divergence between forecasts and actuals for both models with the actual data remaining at very low levels while the forecasts increased. In the holistic model, the lowered spreads, reduced inflation rate, and S&P 500 gains all drive predictions upwards.

In the targeted model, we witnessed spreads increase in 2022, followed by a decrease in 2023. According to the model, that combination will lead to increasing prepayment rates.

To explain this divergence, we can consider the possibility that pessimistic expectations about the future state of the macroeconomy led borrowers to avoid prepayment. The series of bank failures in March 2023 led to serious concerns of a potential wider banking and economic crisis.

Beyond that, the continued interest rate hikes led to widespread concerns that the effort to beat back inflation might cause a recession. The volatile geopolitical environment further exacerbated these effects to increase market volatility and stress market sentiment. With these different perceived risks to the macroeconomy, borrowers would likely hesitate about prepaying even if the current macroeconomic data showed signs of improvement. 

We can see evidence of forward-looking pessimism in a number of data sources. According to NABE, a majority of Economists surveyed in February 2023, expected a recession in the next 12 months. Furthermore, consumers showed similar pessimism regarding the future.

Within the Consumer Confidence Index, the Present Situation Index and Expectations Index diverged in recent data. The Present Situation Index recovered substantially in 2021 and has remained at a high level. Whereas expectations began dropping in 2022 and have recovered only slightly since then.  However, to fully understand the impact of these perceptions on prepayment behavior, it’s important to examine business perceptions as well. After all, businesses are the actual borrowers in this context. 

The OECD’s Business Confidence Index shows a similar pattern. Business Confidence peaks in early 2022 as it recovers from the COVID pandemic, but then begins declining from there. In early 2023, it bottoms out close to its pre-COVID level and remains at that low level.

Another frequently discussed forward looking indicator for recessions is the 10-year 2-year Treasury Spread. The indicator has predicted every recession between 1955 and 2018. The yield curve has been negative since July 2022, suggesting that markets are expecting a downturn in the new future.

Having considered an explanation for the forecast divergence, one can consider modifying the model to improve its performance going forward. Given the popularity of the 10-year 2-year Treasury spread as an indicator of recessions, we can consider using that in the model.

We take the holistic model consisting of S&P 500 growth, High Yield Spreads, and CPI growth. Both the S&P 500 growth rate and the 10-year 2-year spread are capturing future expectations of performance. So, we can try substituting the S&P 500 growth rate with the 10-year 2-year Treasury Spread.

When we evaluate the new model, we find that it does better in back-testing exercise for 2023 (as see in Exhibit 5), as well as in a rolling back-test exercise where we consider each of the previous years as potential holdout samples. Given that, the revised model seems to do better forecasting both history and recent data, we can consider using it as a champion model going forward.

Exhibit 5: Out-of-sample results for Revised Model

Out-of-sample results for Revised Model

Conclusion

In this article, we have explored how macroeconomic factors drive prepayment risk in the BSL market and the importance of stress testing and scenario analysis in making informed investment decisions. Through our Risk Modeler Analytics platform, we presented two approaches for thinking about macroeconomic factors: 1) a more sophisticated approach that considers numerous macroeconomic factors and 2) a more straightforward approach that is easier to implement for scenario analysis.

We then reviewed the out-of-sample results and recent macroeconomic data. We found that incorporating 10-year 2-year Treasury Spreads as a factor in the model could improve the model’s out-of-sample forecasting results and have added that into a new model moving forward.

In conclusion, understanding how macroeconomic factors affect prepayment risk is crucial in evaluating BSL investments. By utilizing scenario analysis and macro-based models, investors can gain insights into borrower behavior and make informed decisions that drive analysis and decision-making to allow BSL managers to manage their portfolios and optimize their returns.

Fortunately, a holistic model that incorporates inflation, 10-2 Treasury Yield Spread, and high yield spread variables offers a powerful, intuitive, and effective model input to facilitate this important prepayment analysis. Stay tuned for the next prepayment analysis of loans issued by public capital market participants and retail lenders.


Key contacts

Steve Kernytsky

United States

Manager, Quantitative Analytics

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

Variation in BDC portfolios can offer investors opportunities to better manage risk and return


architecture modern curves

Summary

Private credit markets continue to be an important source of funding within the leveraged loan universe. This is especially the case for smaller to middle-sized companies, those with EBITDA of roughly $10mln-$250mln. Non-bank lenders have emerged as a growing source to funding to these borrowers through a variety of platforms.

Examples of these alternatives include direct lenders, closed-end funds, to a lesser extent, collateralized loan obligations (CLOs), which generally fund larger and broadly syndicated loans (BSL), and last but not least, business development companies (BDCs). We have recently turned our analytical attention to  BDCs as they provide both institutional and retail investors an opportunity to participate in the small and middle market lending asset class and as many are publicly traded.

Although BDCs typically invest in senior secured loans to smaller and middle-sized U.S. borrowers, the portfolio compositions can vary in many dimensions across BDCs, and in some cases quite significantly. Our analysis has highlighted key elements within the portfolios that would indicate the degree of correlation and hence effect the variance in the performance of a BDC’s returns (or losses) in two dimensions – issuers and industry.

These differences are particularly important for BDC investors (or lenders) as performance metrics would be further magnified due to the leverage that BDCs employ to finance themselves, similarly to CLOs, many closed-end funds and other leveraged investment vehicles. In essence, the relative risk/return profiles for these market participants can be more exposed to “fatter-tails” in the distribution of returns of the underlying portfolio due to less diversification (i.e., potential for greater volatility) – or conversely smoother and more predictable risk/return profiles due to greater diversification within the BDC.

Key performance drivers are related to the overall portfolio diversification of the BDC based on its concentrated exposure of investments made to unique borrower/issuers and industries/sectors, which are the focus of this paper, amongst other factors. We demonstrate through observations, which are based on publicly available information, that clearly indicate significant differences of diversification across BDC portfolios. However, this variation of portfolio composition also provides investors (or lenders) opportunities to better manage across the risk/return and portfolio diversification spectrums.

BDCs – brief overview

The US Congress initially established BDCs in the 1980s to offer an alternative source of funding to developing smaller and middle-sized companies. BDCs are generally subject to the same rules and regulations that apply to US regulated mutual funds – the Investment Company Act of 1940 and the Internal Revenue Code of 1986 for tax purposes.

BDCs are typically leveraged, subject to a maximum amount, with financing typically based on a combination of credit facilities and term debt. BDCs are subject to certain investment criteria (e.g., minimum percentage of investments consisting of smaller U.S. companies) and are required to pass through at least 90% of earned income.

The underlying portfolios typically comprise of first-lien senior secured loans to private corporate borrowers – other possible investment strategies may be around subordinated loans, mezzanine debt, special situations, and distressed debt, amongst others.

The investments are managed by asset managers with extensive expertise and operations capacity in this asset class where, in many cases, the manager may already participate outside of the BDC as direct lenders, private equity sponsors and/or manage other types of vehicles (e.g., CLOs, private accounts, closed-end funds).

In a previous analysis, we highlighted some key differences between private credit and BSL as well as factors for investors to consider. BDCs are generally substantially weighted towards middle market and smaller company lending although there are some BDCs with notable CLO/BSL exposure.

Those BDCs that are publicly traded not only offer liquidity to investors but also a great amount of transparency. For example, like other public companies, they are required to provide quarterly and annual reports. These reports make available to the public numerous detailed information related to the underlying portfolio and is the primary source of information our observations are based on.

It should be acknowledged that these are point-in-time snapshots of the portfolio and could change over time as BDCs are going concerns and actively managed. However, they do provide valuable insight into the manager’s lending/investment philosophy, and we believe our conclusions and findings are generally valid across time. Market participants should view this as an example of portfolio variation across the public BDC spectrum at a given point-in-time.

Comparing diversification across BDCs – key observations

We selected a sample of 32 BDCs to compare amongst the 40+ that are publicly traded (see Appendix). These examples were chosen to clearly demonstrate how the composition of the portfolios can vary significantly across BDCs. In this case, the focus is with respect to investment diversification as reflected by the distribution of borrower/issuer and industry/sector concentration levels. Other diversification benchmarks, which is outside the scope of this paper, could include investment strategy, credit risk, geography, currency, seniority of debt, asset class, level of affiliation, etc.

See Exhibits 1-4 below illustrating various concentration statistics followed by noteworthy observations. Note that the percentages and amounts of the respective portfolios are based on fair market value in USD as reported, in most cases as of the most recent calendar year-end. In addition, we excluded any investments that were in cash (or its equivalents) or had a fair market value of zero.

Borrower/issuer concentration

Exhibit 1: Top Seven Borrower/Issuer Concentrations vs. Average Concentration

The distribution of credit and investment exposures to unique borrowers/issuers is perhaps the best indicator to begin an assessment of the level of portfolio diversification. A BDC manager may allocate several investments to a single borrower/issuer, for example, based on a combination of loans/debt and/or equity. As such, we combined all investments to a unique borrower/issuer for purposes of determining concentration. We noted that on average there were around 1.5 investments per unique issuer with ranges of close to 1 and a third of our sample in excess of 2.0.

In many instances, portfolios tend to be fairly concentrated within the largest few exposures, but relatively granular with the remainder. See Exhibit 1 above illustrating that the largest seven borrowers/issuers range from approximately 10% to about half of a given portfolio.

Also, BDC portfolios can consist of less than 30 investments allocated to unique borrowers/issuers to more than 300. The exhibit also illustrates the level of average borrower/issuer concentration in the portfolio that provides additional insight into the degree of granularity across the portfolio.

It is important to note that, in several BDCs, aside from significant exposures to general corporate borrowers, some of the largest investments are allocated to pools of investments (or  ‘funds’). These funds can consist of an underlying portfolio of credits financing private corporate borrowers, real-estate, or equipment as examples.

They can be managed by the same manager of the BDC, a joint venture with the BDC manager, or another fund altogether (e.g., CLO, another BDC, closed-end fund). The latter would likely be managed by a separate entity.

In some cases, these funds can be around 10% or even greater than 20% of the BDC portfolio. Obviously, one can argue that these investment allocations indicate more diversification than one made to a single corporate borrower.

However, the degree of diversification would depend on exactly what investments underly the fund and its individual exposures versus the rest of the BDC portfolio (e.g., borrower overlap). In addition, for an investment in a fund that is levered, like a CLO, it would also depend on what seniority position the investment is within the capital structure (e.g., senior, mezzanine, equity).

Industry/sector concentration

Comparing industry/sector exposures across BDC portfolios is a bit of a challenge since there is no standardized industry reporting classification taxonomy used across the BDC market. Some BDCs classify industries more broadly across their investments while others classify them on a more granular level.

To further complicate matters, as mentioned earlier, certain BDCs have investments that are in funds. It is more difficult to determine industry concentrations in those cases unless one can look through to the underlying investments within those portfolios. For example, it may be unlikely that a CLO portfolio has any significant borrower overlap with the rest of the BDC portfolio since they typically comprise of BSL and that they are often managed independently.

However, this may not necessarily be the case with respect to industries/sectors. In which case, market participants would need to rely on their own assessment based on information made available. Nonetheless, determining diversification across funds or CLO portfolios within a BDC portfolio is beyond the scope of this paper as it can depend on many factors (comparable to those presented here).

Therefore, to make the BDCs more comparable in this respect, with judgement, we mapped the industry classifications that each BDC reported for each investment to a standard group of industry categories.

Exhibit 2: Top Three Industry/Sector Concentration vs. Average Concentration

Initial observations indicate that BDC portfolios are generally concentrated amongst several industries. See Exhibit 2 above illustrating that the largest three exposures typically exceed 40% and in some cases exceed 70% of the overall portfolio. Additionally, the total number of industries represented across a portfolio range from less than 10 to more than 20.

Again, note our earlier comments as some of the top industry classifications may include investments in funds consisting of underlying portfolios. As with Exhibit 1, the exhibit also illustrates the level of average industry/sector concentration in the portfolio providing additional insight as to  the granularity across the portfolio.

Exhibit 3: Common Industry/Sector Concentrations Across BDCs

It is also noteworthy that several industries commonly appear across the BDCs (see Exhibit 3 above). Specifically, these industries are related to high-tech, healthcare & pharmaceuticals, business services, and finance. Furthermore, in many cases, these industries are within the top three most concentrated (see Exhibit 2).

This is a particularly important factor for those investors (or lenders) that have exposures to a portfolio of BDCs when assessing the correlation across their investments as these industries represent some of the largest investments across the universe of BDC portfolios, in terms of both the number of individual investments and fair market value.

Of note, however, these industries could be diversified in their own right into finer industries that would imply lower correlation – for example, within ‘high-tech’ there are companies that focus on educational software, construction technology, cyber, electronics, green technology, health-tech, manufacturing technology, etc.

Not surprisingly, however, the most concentrated BDC portfolios can be less correlated with others since they may have minimal exposure to (or exclude) many industries. This is obviously another key factor for those market participants that allocate (or monitor) investments across BDCs as it can provide an opportunity for additional diversification and/or targeted allocations to certain industries in the private markets.

Asset type/seniority concentration

While we won’t delve into detail for this article, but may be explored further in future Alter Domus research, it is worthwhile noting that the type of and/or seniority level of the investments can vary across BDCs. See Exhibit 4 below for concentration levels across various asset classes as reported by the BDC. We grouped certain types together as they can be considered as other and with relatively small exposures with the exception of CLOs (particularly the most subordinated), which can be significant.

Exhibit 4: Asset type/seniority concentration

Exhibit 4 shows that debt is commonly the most significant portion of the portfolios. This is to be expected as BDCs typically extend credit to corporate borrowers usually in the form of senior secured loans. Within the debt spectrum, these can vary from senior secured loans that are on a first or second-lien basis to those that are unsecured or in the form of notes.

As mentioned earlier, BDCs can offer financing to a single borrower across the capital structure (e.g., senior secured, mezzanine, equity) depending on the BDC manager’s financing strategy. The combination may explain some of the significant equity exposure. The equity exposure can be an investment in a levered fund as well.

Conclusion

Non-bank lenders, within the private credit market, have been an important alternative source of loan funding for smaller to middle-sized U.S. companies through various platforms (e.g., direct lenders, closed-end funds, BDCs). BDCs are particularly worthy of attention since they offer a spectrum of investors an opportunity to participate in this lending asset class and as many are publicly traded.

Portfolio compositions can vary quite significantly across BDCs even though BDCs typically consist of leveraged senior secured loans extended to smaller and middle-sized U.S. borrowers. These differences across BDCs offer investors a menu of risk/return profiles to suit their needs.

We presented our observations on variations in diversification across BDC portfolios based on an evaluation of elements that are key indicators to the overall level of diversification, with an emphasis on concentrations to unique borrower/issuers and industries/sectors.

In conclusion, our analysis indicates significant differences of diversification across BDC portfolios, which provides investors opportunities to better manage across the risk/return and portfolio diversification spectrums.


Appendix: List of BDCs

BDC NameTrading Symbol
Ares Capital CorpARCC
FS KKR Capital CorpFSK
Blue Owl Capital CorpOBDC
Blackstone Secured Lending FundBXSL
Prospect Capital CorpPSEC
Golub Capital BDC, IncHTGC
Main Street Capital CorpMAIN
Hercules Capital IncHTGC
Sixth Street Specialty LendingTSLX
Oaktree Specialty LendingOCSL
Barings BDC, IncBBDC
MidCap Financial Investment CorpMFIC
Bain Capital Specialty Finance, IncBCSF
SLR Investment Corp.SLRC
CION Investment CorpCION
BlackRock TCP Capital Corp.TCPC
Crescent Capital BDC Inc.CCAP
PennantPark Investment Corp.PNNT
PennantPark Floating RatePFLT
Trinity Capital IncTRIN
Capital Southwest CorporationCSWC
Fidus Investment CorpFDUS
Stellus Capital Investment CorpSCM
Saratoga Investment CorpSAR
Gladstone Investment CorpGAIN
Horizon Technology Finance CorpHRZN
Portman Ridge Finance CorpPTMN
Monroe Capital MRCC
Oxford Square Capital CorpQXSQ
Great Elm Capital CorpGECC
Loan Ridge Finance CorpLRCF
PhenixFin CorporationPFX

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

AI & Alternatives case study: Consolidating files into a NAV dashboard

We at Alter Domus are regularly approached by our key clients to help them solve some unique business challenges. As both our expertise and investment technology have grown significantly, we have increasingly begun to utilize artificial intelligence as a central mechanism to find and deliver solutions to our clients. This is the first in a series of case studies designed to highlight the multiple ways we are creating impactful, customer-centric technology for alternative markets.


technology close up data on screen scaled

The challenge

Getting a clear picture of your net asset value, or “NAV” as it’s known, is a crucial calculation for every investment company. Essentially, a company’s NAV is its total assets minus its total liabilities, and as the number of funds and assets accumulate, the task of calculating this becomes ever more arduous – especially when using legacy tools such as Excel.  We were approached by one of our key clients to see if we could create a solution to this labor-intensive task.

Their team faced the challenge of having to manually consolidate up to 11 Excel files to create one final NAV dashboard for each of their funds. What is of course crucial is that as the volume of data and manual work increases, the greater the possibility for data inaccuracies and calculation errors. On average, this entire process took days for their team to complete.

The solution

To address this challenge, the team proposed and designed a solution that involves the user uploading the 11 Excel files, including trial balances, bank balances, and FX rates, into a tool that consolidates them into a single NAV dashboard. Once the consolidation is complete, the user receives an email containing the new dashboard.

How the process and AI bot works (per each fund)

STEP 1:

  • Alter Domus team logs into the Web Apps Portal and runs our NAV tool

STEP 2:

Alter Domus team uploads the 11 Excel sheets received from client, including:

  • Trail balance, bank balance, collateral abacus & eFront
  • FX rates, loan & credit facilities, accruals listing
  • Loan request, ICAS previous & current period, hedging

STEP 3:

  • Bot opens each Excel sheet. It copies and pastes, and completes lookups from the column data in the various spreadsheets
  • Bot cleans data where required

STEP 4:

  • Significant reduction of the time and effort required by client’s team to complete this task: what once took days now takes minutes
  • Enhanced data and calculation accuracy
  • Client’s team is freed up to focus on less manually intensive, more strategic tasks
  • The bot is able to handle increasing volumes of this work, eliminating the need for the client to hire additional staff or reallocate internal resources 

Implementation time

  • 2 months to develop the bot
  • 1 month in UAT
  • 1 month in control production

Contact us today to find out how you, too, could benefit from our use of artificial intelligence in delivering impactful technology solutions.

Key contacts

Davendra Patel

Davendra Patel

Europe

Head of AI & Automation

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

Post-LIBOR transition update

Calm after the storm


architecture bridge traffic

After more than 36 years of existence, the London Interbank Offered Rate (LIBOR) was discontinued in July 2023 in favor of alternative references rates, such as SOFR (Secured Overnight Financing Rate). As mentioned in previous Alter Domus reports, we discussed the need to migrate away from LIBOR and provided updates on the transition as the cessation date grew closer.

Now that LIBOR is no longer published, we report on the impact it continues to have, albeit declining, in the U.S private debt market and address any lingering considerations.

Announcement of LIBOR’s future discontinuation in 2017 posed an intimidating feat, like an impending storm, as LIBOR was pegged to almost all contracts accounting for billions of dollars of debt. Six years later, though, market participants made it through the storm quite successfully.

Overall, the LIBOR transition was, as the AARC[1] puts it, “smooth and uneventful.”[2] Much of the success is attributed to the careful planning and coordination between agents, like Alter Domus. borrowers, lenders and financial regulators.

Moreover, as the June 30,2023 deadline approached, there was a predictable uptick in credit agreement amendments and triggering of existing ones. As reported by the LevFin Insights, amendments spiked in June, nearly totaling 300, representing almost 3 times the amount of amendments in May and April[3].

The same can be inferred with respect to the direct lending space, as we turn to publicly available BDC (Business Development Companies) data as a proxy. From a sample of 13 BDC’s registered with the SEC, representing over $33bn in portfolio valuations or about a quarter of the publicly traded BDC market, we observe from the 2023Q2 SEC filings that the average decrease in loans tied to LIBOR jumped to 27% during 2023Q2 from 10% in 2023Q1.

The graph and table that follow clearly illustrate the declining relevance of LIBOR and rising relevance of SOFR in the last four quarters – with SOFR beginning to equal and surpass LIBOR in Q1’23. Over the past 9 months, SOFR-pegged loans have gone from 28.5% of total floating loans to 75% for the loans in the underlying sample portfolios.

LIBOR Declines While SOFR Grows (sample of 13 BDCs)

Count of Floating-Rate Investments – Sample of 13 BDCs over the past 4 quarters

Base RateQ3’22Q4’22Q1’23Q2’23
LIBOR-based floaters 1,2321,117986454
Non-LIBOR-based floaters5507691,0051,555
Total Floaters 1,7821,8861,9912,009
LIBOR as a % of Total Floaters69.1%59.2%49.5%22.6%
SOFR-based floaters5077159421507
SOFR as a % of Non-LIBOR92.2%93.0%93.7%96.9%
SOFR as % of Total Floaters28.5%37.9%47.3%75.0%

We’ve currently discussed the recent increased migration efforts that led to a successful migration away from LIBOR, but does that mean all loans refer to an alternative rate? Not quite. Currently, in the same BDC sample as of the end of 2023Q2, 75% of loans reference SOFR.

With regards to the leveraged loans and the CLO market, the results are similar, as 76% of loans reference SOFR (as of August), per LSTA.[4] The reason why these numbers are not 100% is due to contracts that were opened prior to the June 30,2023 deadline that still refer to a LIBOR reference rate.

Since most contracts are 3-months, we should expect to see a considerable amount of loans automatically migrating to SOFR (via the LIBOR act or through the triggering of an amendment). We will continue to track this migration as it nears completion.

Overall, it appears that the transition has gone quite smoothly, thanks to years of thoughtful preparation. Loans that are currently tied to LIBOR will eventually point to an alternative rate once the contracts next renew – post June 30.

The AARC, which was tasked to determine the proper alternative rates to transition to, will slowly[5] wind down its work, emphasizing that the SOFR should be deemed the best option as the alternative rate given its robust tie to the transaction-heavy treasury market.

Going forward, we will keep readers updated on the status of the remaining loans that still reference LIBOR and hope to look past the migration with a greater sense of triumph.

Sample of BDCs in our study:

Audax Credit BDC Inc.
BlackRock Private Credit Fund
BlackRock TCP Capital Corp.
Capital Southwest Corp.
CION Investment Corp.
Fidus Investment Corp.
Great Elm Capital Corp.
Oaktree Strategic Credit Fund
Oaktree Specialty Lending Corp
Blue Owl Capital Corp.
Oxford Square Capital Corp.
PennantPark Floating Rate Capital Ltd.
Prospect Capital Corp.

[1] Alternative Reference Rates Committee

[2] https://www.newyorkfed.org/medialibrary/Microsites/arrc/files/2023/ARRC-Readout-July-31-2023-Meeting

[3] https://www.lsta.org/news-resources/libor-transition-one-two-last-hurrahs/

[4] https://www.lsta.org/news-resources/libor-cessation-t1-month/

[5] https://www.newyorkfed.org/medialibrary/Microsites/arrc/files/2023/ARRC-Readout-July-31-2023-Meeting

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

Understanding the impact of excess interest on CLO portfolios


technology close up data on screen scaled

Alter Domus has previously demonstrated through a simple example that total credit losses from an underlying collateralized loan obligation (CLO) portfolio would equal to the total credit losses across the CLO’s notes (including equity). In that same research Alter Domus pointed out, however, that CLOs typically include structural features where excess interest, which would otherwise have been distributed in traditional portfolios, such as bank balance sheets or insurance company general accounts, can be diverted to provide credit support for CLO tranches upon an economic downturn.

This diversion of excess interest income means that total portfolio credit losses are reduced by the amount of excess interest diverted. In particular, the formula in estimating CLO tranche credit losses, as a function of the portfolio credit losses, is: Losses Allocated to CLO Tranches = Total Portfolio Credit Losses – Excess Interest

In this article, we expand our previous research by conducting an analysis to assess the impact of the relative benefit of excess interest, which is highly dependent on the overcollateralization (OC) test, as it translates to the reduction in the expected loss (EL) of the underlying loan portfolio.

Specifically, the purpose of our research is to estimate the ‘Excess Interest’ from the formula above. As expected, we find that excess interest can mitigate credit losses in the underlying portfolio, but the extent to which is sensitive to several variables, such as the average credit quality of the portfolio, the available amount of excess interest over the life of the CLO, and portfolio weighted average life (WAL).

The results of our research indicate that a reasonable range of the reduction in portfolio EL can be from 6%-14%, with an 11% average, depending on the level of ‘stress’ one wants to apply in their analysis. Other factors, such as prevailing market conditions, could also impact this range. Based on our findings, we could rewrite the formula above as follows:

  • Low Value: Losses Allocated to CLO Tranches = 86%* Total Portfolio Credit Losses 
  • High Value: Losses Allocated to CLO Tranches = 94%* Total Portfolio Credit Losses 
  • Average Loss: Losses Allocated to CLO Tranches = 89%* Total Portfolio Credit Losses 

A review of the OC test

CLO market observers understand that a critical feature in CLOs is the benefit of excess interest generated by the underlying CLO portfolio to provide credit support for the CLO tranches. But the benefit of excess interest in reducing the EL for CLO tranches is highly dependent on several variables.

A critical structural feature in CLOs that helps ‘trap’ this excess interest to support the CLO tranches is the OC test (note that CLO structures will typically have several OC tests at various points in the waterfall – in this paper the singular term ‘OC Test’ refers to all OC Tests collectively).

With no OC Test, the benefit of excess interest will be limited because most of the excess interest will leak out to the equity investors – this would mirror to a certain extent the treatment of excess interest for a portfolio of loans on a bank’s balance sheet or an insurance company’s general account, where the income could be used for other purposes by those institutions and not trapped and secured for the benefit of the CLO noteholders.

OC tests are generally incorporated within the CLO cashflow waterfall. Upon a breach of such test, they are intended to divert available interest proceeds first, then principal proceeds if needed, to pay down the notes in order of priority until cured. Typically, there is a test for each class of notes except for the class that was initially the senior most. In that case, any proceeds diverted are at the expense of the equity including the relevant junior notes depending on which test is breached.

The OC test is usually calculated based on the ratio of the total par amount of the underlying portfolio (plus principal proceeds) over the outstanding amount of the relevant class of notes (including those that are senior to it). However, there are adjustments in the form of haircuts that are commonly applied to the portfolio.

These typically include those that are classified to be in default or purchased at a deep discount as well as the total amount of those rated Caa/CCC in excess of a predefined limit. OC tests are more likely to be breached upon credit losses and/or excessive negative credit migration.

It is important to note however that the ultimate benefit of this feature can depend on many factors as CLOs are complex and the behavior of the economic environment is uncertain. Some examples of these factors include:

  • Available excess interest (e.g., weighted average spread of underlying loans, cost of debt across the CLO notes, behavior of reference rates)
  • Amortization profile, prepayments (i.e., introduces reinvestment risk)
  • Timing of defaults (e.g., front loaded defaults may trigger tests earlier)
  • ‘Tightness’ of OC test trigger levels (e.g., lower trigger levels may reduce effectiveness of tests)
  • Portfolio manager trading behavior (e.g., can increase – or decrease – par amount of portfolio through substitutions particularly when an OC test is failing marginally)

Assessing the benefit of the OC tests – a simple analysis 

We have just described the basic mechanics of how OC tests work and highlighted a few examples that can impact their benefit, several of which are quite uncertain and challenging to forecast. Therefore, we choose to control for most of those variables and start from a simple stylized CLO as the intention of our analysis is to demonstrate the economic benefit of excess interest.

The benefit from this form of potential credit enhancement can be explained as to the reduction of the underlying portfolio EL. The analysis will also include an examination of the sensitivity while changing several variables that are straightforward.

To begin, we created a stylized CLO by reviewing a sample of recent CLO structures. The objective of this step is simply to assess how much excess interest is generated in a CLO structure. However, it is safe to say that excess interest is a prerequisite for any CLO structure.

Although the amount of excess interest would change over time based on market conditions, we could generalize that a CLO structure would not be issued when market conditions do not offer sufficient excess interest. See Tables 1-3 for assumptions made for our hypothetical CLO.

As we had noted in previous research, the collateral quality test matrix, typically incorporated in CLOs, is a clear indicator of the benefit of excess interest – the EL of the CLO notes would not be expected to materially change even as the credit risk of the portfolio increases because of the reduction in losses resulting, for example and most commonly, from an increase in excess interest.

The Matrix Scenarios provided in Table 3 were calibrated so that the tranche ELs of our hypothetical CLO were relatively consistent in each of the designed scenarios. CLO matrices typically provide many more options than the modeled scenarios presented here.

The portfolio is assumed to be static, consisting entirely of floating rate first-lien senior secured loans. Recoveries are also assumed to be realized immediately upon a default.

Table 1: Hypothetical Capital Structure

Table 2: Subset of Collateral Quality Tests

Table 3: Collateral Quality Test Matrix

With this stylized CLO, we assumed that the reference rate is 0%. This assumption allowed us to get a clearer picture of the excess spread that could be applied to reduce the portfolio EL. As reference rates rise, the amount of excess interest, depending on a given scenario of losses, can increase thus potentially offering greater credit enhancement.

For our analysis, we ran our model for each matrix scenario under a range of different stresses. We also ran various combinations of default rates, recovery rates and portfolio WAL for each of the matrix scenarios. The results of the various combinations are shown in Tables 4 and 5.

For each of the four matrix combinations, we ran the CLO cash flows assuming various levels of default rates (historical averages based on ‘idealized’ levels) and stressed default rates. We also ran two different recovery rate scenarios – again stressed and closer to historical averages.

We also ran two combinations of portfolio WAL. This broad range of assumptions allowed us to study the sensitivity of the excess interest in reducing portfolio EL. The column labeled ‘% Reduction of Portfolio EL – Note Paydowns Only’ represents the probability-weighted amount of interest proceeds that are diverted to pay down the notes, excluding any payments to the first-loss equity tranche. We believe that this number is a reasonable estimate of the portfolio EL reduction attributed to excess interest.

Table 4: Portfolio WAL of 7 years

As we mentioned earlier, the collateral quality matrix offers the portfolio manager to choose amongst different combinations of portfolio characteristics (e.g., average rating, WAS, diversification) while maintaining the credit quality of the notes. In other words, and as we had noted in our previous research, the collateral quality matrix is the proof and our guide in trying to assess the reduction in portfolio EL resulting from excess interest.

In Table 4 we note that for the four different scenarios, the average reduction in Portfolio EL attributed to the excess interest ranges from 6.4% to 17.1%. As expected, the percent reduction of the portfolio EL increases as the portfolio EL increases within each set of matrix scenarios. Indeed, this result is also indicated by the percent of the total interest proceeds that is diverted to pay down the notes due to the OC test. In the matrix scenarios presented here, the ‘required’ WAS, an indicator of excess interest, of the portfolio increases as the average credit rating decreases.

It is also worthwhile noting that the proportion of excess interest that is distributed between the equity and what is diverted due to the OC Test over the life of a CLO is subject to uncertainty, some examples of which we highlighted earlier. However, it can be said that holding all else equal, a greater proportion would be expected to be diverted, the higher the OC trigger level. In which case, it would lead to a further enhancement against the portfolio EL.

Table 5: Portfolio WAL of 5 years

Comparing across the results within Table 5 show a similar behavior as in Table 4. However, when comparing the results between the two, notice that both the percent reduction to the EL and the percent of the total interest proceeds that are diverted are lower. This can generally be explained by a combination of lower portfolio ELs from lower probabilities of default and that the expected amount of available excess interest over the life of the CLO is lower, both due to the shorter duration of the portfolio. The range of portfolio EL reduction due to excess interest is 5.1%-10%.

Nonetheless, the results from the tables above are compelling in demonstrating the case that excess interest reduces the portfolio EL and provides our first estimate of the ‘Excess Interest’ in terms of reduction of the portfolio EL.

Based on the assumptions above, and for the broad range of different model inputs, we found that excess interest can reduce the portfolio EL in the range of approximately 6%-14% with an average portfolio EL reduction of about 11%. We derived this range by simply averaging and rounding the ranges of the two different tables.

We have also shown that it is quite challenging to derive a precise point estimate of the benefit of excess interest given the complexity of the analysis, the various assumptions that could be made and changing market conditions. As a result, we provide our views of a reasonable range of the benefit of excess interest in reducing portfolio losses. Essentially, we attempt to answer the following question:

‘What is the benefit from Excess Interest based on the formula below?’

Losses Allocated to CLO Tranches = Total Portfolio Credit Losses – Excess Interest

We believe that, based on our example, a reasonable range could be somewhere from 86% to 94% of the gross portfolio EL as compared to assuming no excess interest in the CLO. 

A few notes

We have noted that our analysis included certain assumptions in order to demonstrate the impact of excess interest in reducing portfolio EL. Beyond the assumptions that have already been noted, other considerations that could impact our results include:

a. We assumed that none of the equity distributions reduce portfolio EL. This is a conservative assumption but given the complexity in trying to  estimate the composition of equity distributions between interest and principal are beyond the scope of this paper. Either way, the equity tranche assumes the bulk of the portfolio EL anyway – as demonstrated in our previous research paper.

b. Our hypothetical CLO – see Tables 1-3 – was created based on relatively recent market conditions. Hence, the relatively large spreads on the notes (and to a certain extent, the portfolio). Under different market conditions, the results of our findings could be different. But generally, with narrower spreads on the notes, the benefit of excess interest increases and thus further reducing portfolio EL.

c. We used a limited number of matrix combinations. Using a broader and finer range of matrix combinations may show different results under different collateral quality test combinations.

Conclusion

In this article, we have assessed and demonstrated the credit enhancing benefit of excess interest within CLOs as they typically incorporate a key structural feature, the OC test. In the event the OC test is breached, excess interest, which would have been distributed to equity, would be available to pay down the most senior note until cured.

Our analysis was based on several assumptions applied to a simple hypothetical CLO structure. Several scenarios were applied to observe the relative impact while changing several variables: the average credit quality of the portfolio, the available amount of excess interest over the life of the CLO, and the portfolio WAL.

The metric used to measure the relative benefit was based on the reduction in the EL of the underlying loan portfolio. As expected, we found that excess interest proceeds generated from the portfolio reduces the EL on the CLO tranches in all cases.

The results show an average reduction in portfolio EL of about 11%, though this amount is subject to a limited range of assumptions.

Although the results indicate that the impact can be sensitive, it is nonetheless clear and not surprising that this key feature of trapping excess interest, at a minimum, can provide additional credit enhancement.

Key contacts

Rudolph Bunja

United States

Head of Portfolio Credit Risk

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference

Analysis

LIBOR cessation in less than 30 days: sound the alarm?


man in boardroom staring out of window

Exploring options and best practices

The London Interbank Offered Rate (LIBOR), a controversial cornerstone of the financial system, will be discontinued on June 30,2023. In a previous article published in December 2022, we discussed the need to migrate away from LIBOR and the migration status for the U.S private debt market[1].

As LIBOR will be discontinued in less than 30 days, we will provide updates on the current migration status and answer key questions: what happens if no explicit migration plan exists before LIBOR is discontinued and what are the current industry trends regarding credit adjustment spreads[2]?

Current Status of migration for U.S. Private Debt Market

According to the LSTA, in reference to the JPM index as of April 2023, roughly 40% of private loans are currently on SOFR[3]. With respect to the direct lending space, we turn to publicly available BDC (Business Development Companies) data as a proxy. From a sample of 12 BDC’s registered with the SEC, we observe from the 2023Q1 SEC filings that on average 49% of loans are pegged to SOFR.

For the same sample, 2022Q3 and 2022Q4 SEC filings report SOFR loan percentages of 23% and 35%, respectively, showing an increased effort to transition away from LIBOR as the cessation date approaches. While industry trends suggest a concerted effort to convert loans to SOFR, we would likely to expect these figures to be close to 100% as we have to count days and not months until LIBOR is discontinued.

However, data from Covenant Review reveals that over 90% of syndicated loans have language in their contracts or amendments that determine the alternative rates at the time of LIBOR cessation[4]. Moreover, as discussed in the previous paper, industry trends point in the larger private debt space point to a larger effort to add amendments to contracts to trigger immediatley after LIBOR cessation date rather than transition earlier. Therefore, although the numbers are not nearly close to 100% as we would expect, there are already contractual agreements in place for a vast majority of loans to migrate away from LIBOR once it is discontinued.

But what about the contracts that are not in the “vast majority” that do not have an alternative rate in place come June 30,2023? What will those loans be tied to if not LIBOR? The Federal Reserve Board, in December 2022, provided its final set of guidelines for such contracts[5].

Fallback options for contacts without alternative rate transition plans prior to June 30, 2023

Final rule regarding the implementation of LIBOR ACT

Widely expected, the Federal Reserve Board in December 2022 published its final guidance to implement the LIBOR act[6]. This guidance is to explicitly provide a benchmark replacement for contracts governed by U.S law that refer to either the overnight, 1-month,3-month, 6-month and 12-month LIBOR tenors and do not have a clearly defined benchmark to transition to after June 30,2023. 

In general, the guidance determines if there is a “determining person” in the contract. If so, that person should determine the replacement contract by the LIBOR cessation date. If the person does not, or if there is no “determining person” in the contract, the contract must transition to the corresponding Term SOFR plus the prescribed credit adjustment spread based on ISDA/AARC fallback protocols[7][8].

As Term SOFR is typically lower than LIBOR (as it does not involve credit risk), the credit adjustment spread is added to the current spread/margin to reflect a number closer to LIBOR. For clarity, if a loan has locked in a 6-month LIBOR rate on June 15, 2023, it will still be valid until the contract ends on December 15, 2023. After the contract ends on December 15, however, the loan must be tied to an alternative rate. The table below summarizes the final guidance for the current published LIBOR tenors:

In short, if there is no transition plan, the guidance will provide one. The guidance provides a particular credit adjustment spread depending on the tenor (1-Month, 2-Month, etc.), but how were those spreads determined and how are market participants adapting to them?

Credit Adjustment Spreads: ISDA/AARC Recommendations and Trends

Because private debt loan amounts generally range between 10s or 100s of millions of dollars, it is no surprise that both borrower and lenders are highly sensitive to change in spreads and thus haggle to achieve a mutual agreed upon credit adjustment spread[9]. For example, in a loan of $100,000,000, an increase in 20 basis points to the spread results in an increase of roughly $50,000 in interest payments every quarter. Given a tense negotiating environment, are borrowers and lenders trying to negotiate their own credit adjustment spreads or adopt the ISDA/AARC recommendations? Recent evidence points to the latter.

According to the LSTA, 65% of tracked amendment fallbacks used the recommended credit adjustment spreads in April 2023, up from roughly 57% in March 2023[10]. Although there is no clear trend when looking at January and February, it is likely that the strong percentage will persist in May as the transition deadline approaches and the room for negotiation narrows.

Because the recommended spreads are 5-year median differences (March 2016-March 2021), the values should in theory be stable and subject to less litigation in the near future[11].

Conclusion

As the LIBOR cessation date nears, loan migration away from LIBOR picked up considerably in 2023. Thanks to the final guidelines published by the Federal Reserve Board in December 2022, there is clear guidance to ensure a (relatively) smooth transition away from LIBOR.

It is up to market participants to act proactively to ensure the proposed credit adjustment spreads are economically viable and confirm the final transition plans for each contract. As LIBOR is currently pegged to trillions of dollars, it is paramount for market participants to adequately forecast the financial impacts resulting from LIBOR migration.


Footnotes:

[1]For purposes of this paper, ‘private debt’ refers to broadly syndicated loans and private direct lending markets. See existing Alter Domus report for further details.
[2]The credit adjustment spread added to the current spread/margin on a loan to compensate the lender for switching to an alternative rate, which is typically lower than LIBOR.
[3]https://www.lsta.org/news-resources/libor-transition-45-days/
[4]https://www.lsta.org/news-resources/usd-synthetic-libor-exactly-as-expected/
[5]https://www.federalreserve.gov…
[6]https://www.congress.gov/bill/…
[7]https://www.govinfo.gov/conten…
[8]The credit adjustment spread added to the current spread/margin on a loan to compensate the lender for switching to an alternative rate, which is typically lower than LIBOR.
[9]https://www.bloomberg.com/prof…
[10]https://www.lsta.org/news-reso…
[11]https://www.newyorkfed.org/med..

Key contacts

Jason Mendoza

Jason Mendoza

United States

Senior Quant Modeler

Insights

technology data reflected on spectacles scaled
AnalysisApril 17, 2024

Private capital’s technology tipping point

group at event
EventsApril 23-25, 2024

NAGGL Spring Conference