What this covers

Bryan Krug of Artisan walks through how he approaches high yield bonds and leveraged loans, which he places between investment grade and equities: more economic sensitivity than the former, far less volatility than the latter. He cites aggregate leverage of about 4.25x debt to EBITDA against a roughly 12x multiple implied for S&P midcap equities. The conversation contrasts credit analysts, who ask whether a business is defensible and can delever, with equity analysts, who track growth and earnings beats. It also covers how his research method moved from management relationships, to roughly 150 calls a year with competitors, former employees and suppliers, to proprietary data. One example is tracking hailstorms by zip code against Beacon Roofing locations. He calls this progression Edge Evolution Stages.

The range is wide. Krug criticizes debt-weighted benchmarks, explains how a spread breaks down into expected default loss and liquidity premium, and recounts the 2016 energy rout, when bonds seen as well covered by assets fell from par to 35 to 40 cents. He is sharply critical of rating agencies, citing AIG and AAA structured credit. He describes forced selling from rating downgrades as a pool of technical mispricing, with about $5 trillion of BBB debt in play.

The more contested turns include the erosion of maintenance covenants (85 to 90 percent covenant-lite) and value-stripping at J.Crew, Caesars and PetSmart. He also argues that credit leads equities, that active managers can beat passive in high yield, and that light LBO issuance alongside undeployed private equity capital could signal a large coming debt wave.

Sharpest takeaway

Krug argues high yield credit rewards cash-flow-based underwriting, independent and data-driven research, and disciplined enterprise-value protection, because ratings agencies, covenant erosion, and ratings-driven forced selling create mispricing that active managers can exploit.

  • Asset value at par failed to protect investors in the 2016 energy downturn, while cash-flow and deleveraging capacity mattered more
  • Ratings agencies have been wrong on high-stakes credits (AIG, structured credit), creating inefficiencies in pricing
  • Edge has evolved from management relationships (pre-Reg FD) to independent triangulation to proprietary data, and most credit investors are still early in that evolution

The argument · threads13 threads · 68 claims
0.85

Independent triangulated research, not management access, uncovers mispricing that presentations and screens miss.

14 pointscentrality 5/5
  • Krug's team tracked every hailstorm by zip code and overlaid Beacon Roofing's locations within a 20-mile radius to estimate demand, believing storm activity would beat market expectations, a proprietary edge the market did not price.

    “we actually track every storm that's out there, and every hailstorm and rebel to quantify it by zip code”

  • Krug's edge evolved in three stages: pre-Regulation FD, when management relationships gave early information; post-2001, independent triangulation through about 150 calls a year with competitors, former employees, and suppliers; and now early use of data and analytics.

    “the edge before, used to be having a good relationship with management”

  • Every industry has readily available data that most people do not realize exists, such as census data and industry association shipment statistics, which can reveal product penetration trends over years.

    “Every industry has got readily available data out there, but there's a lot of data out there. I don't think people realize exists.”

  • A purely quantitative approach that screens on debt-to-EBITDA and spreads would have underpriced cyclicality in sectors like autos, which were the biggest high yield sector in 2005 and 2006, and produced severe impairments when profitability swung.

    “that would have been a very painful experience because you would have under-priced the cyclicality in that piece”

  • Many credit investors, particularly on the credit side, rely on management presentations and roadshows, which Krug sees as a simplistic research process; even the weighted average active manager is not yet at the independent-triangulation stage.

    “I think a lot of investors actually, particularly on the credit side, aren't even at the second phase.”

  • Active managers can outperform passive peers in high yield because the market has about 1,600 issuers and inefficiencies in analysis, unlike most asset classes where active managers lag.

    “this is one of the few spaces where active managers actually beat their passive peers”

  • Idiosyncratic company-specific data, such as box office tracking or census vinyl siding data, provides an edge that is hard to scale across many securities, unlike cross-sectional datasets that are widely available.

    “everything that we do from a data source is really more bespoke”

  • Krug's diligence on a credit combines financial modeling, management meetings, and calls with competitors and former employees to triangulate how an acquisition is integrating, and the firm makes about 150 such independent calls a year.

    “We also talked to competitors, former employees as part to really triangulate how the integration was going into the acquisition.”

  • Krug says that the rating-agency critique is a generalized view and not necessarily a source of advantage, since peers also know ratings are imperfect.

    “Patrick asked whether or not that's a source of advantage”

  • Krug tracks box office data weekly and attributed roughly 30 to 40 percent of a quarter's theater box office to Black Panther, using that breakdown to anticipate what drives performance and what is likely to repeat.

    “we can say it was 30 or 40% of the box.”

  • Credit research requires triangulating management claims through about 150 annual calls with competitors, former employees, and suppliers, because issuers and bankers have incentives to present data favorably and cherry-picked charts are common.

    “we do around 150 calls a year with those types of people that really give us an independent view as to what's going on”

  • The old edge of having a good management relationship and anticipating news ended when Regulation FD took effect in 2001, which forced companies to disclose material information broadly rather than privately to favored investors.

    “Regulation FD was when the companies could basically tell you whatever they wanted... that changed in 2001”

  • Investors are starting to get smarter about cash flow-based credit selection, so the edge is evolving, but the market has not yet fully adopted it.

    “It is starting to evolve a little bit, people are getting smarter on it, for sure.”

  • Active management in credit is becoming more competitive, so managers must continually evolve their research process, because the edge from a given process decays as peers adopt similar methods.

    “if you don't evolve your process, I mean, it's just getting more and more competitive”

0.56

Forced selling and other identifiable investor errors create selective mispricing opportunities.

7 pointscentrality 4/5
  • Retailers such as J.Crew and Neiman, with 35 to over 50 percent online exposure, sold off as online retail fears took hold, though Krug viewed their company missteps as reversible.

    “J.Crew and Neiman, which have meaningful online exposure between 35 and over 50%.”

  • Some investors avoid CCC paper because of mandate limits and fact-sheet optics, but Krug buys lower-quality credits when he believes their metrics and business trajectory will improve, accepting a lower yield than the sub-sector average.

    “we tend to look at where we believe the trajectory of the credit is going”

  • There is about $5 trillion of BBB debt, and the biggest opportunity arises when issuers fall from investment grade to high yield, because investment grade holders such as insurers are forced to sell or reduce exposure to sub-investment-grade paper.

    “there's about $5 trillion of BBB debt, and where the biggest opportunity is, is typically when issuers drop from investment grade to high yield.”

  • Mispricing comes from identifiable pools of investor error, such as forced selling caused by mechanical rules around rating changes, and understanding who is on the other side of a trade and why they are selling is central to identifying opportunities.

    “what are the pools of investor error? You could argue forced selling, because of a dumb rating agency changing a letter is a source of investor error.”

  • Krug says the firm will sometimes ignore rating-based consensus and treat a credit as a turnaround, which requires accepting that investor perception may be wrong for a long period.

    “We don't love the cable business, but we really love the broadband business”

  • A small opportunistic sleeve trades technicals in loans or bonds and event-driven situations such as potential covenant violations, which was limited in the current calm market but adds value through the cycle.

    “We have an opportunistic slave... we'll take advantage of technicals in either the loan or the bond market, or we have an event-driven tribal”

  • When bonds migrate from investment grade to high yield, the selling is purely technical rather than fundamental, because capital rules force investment grade holders to sell, which creates a source of investor error that active managers can exploit.

    “when assets come from investment grade, it's just, it's pure technical. Where you have investors basically have to reduce exposure because of a change of downgrade”

0.52

Buying cyclical, out-of-favor survivors for spread compression across core and opportunistic buckets drives returns.

5 pointscentrality 4/5
  • Krug's spread-tightening strategy buys cyclical, out-of-favor companies that can survive a downturn, earning returns from spread compression when a recovery occurs, with underwriting returns of high single digits to 30 percent IRR depending on market conditions.

    “Spread tightening is an example where we're trying to find a cyclical out of favor company.”

  • Krug says he is all-in or all-out, so if forced to leave high yield he would retire rather than move to another asset class, because he does not want to be half-invested in his work.

    “I'm either I'm out and I'm in, and I'm not someone that's going to be half in”

  • Krug's portfolio is organized into three buckets: core holdings in stable businesses with pricing power such as Charter Communications' broadband, spread-tightening positions in cyclically out-of-favor companies targeting high single to 30 percent IRRs, and a small opportunistic sleeve that captures technical or event-driven dislocations.

    “We internally tend to bucket in three criteria. We look at core... Spread tightening is an example where we're trying to find a cyclical out of favor company.”

  • LBO activity has been unusually light relative to the capital private equity has raised, which Krug attributes partly to challenging buyout economics, and this lull has been the biggest surprise in the market.

    “thus far it's been relatively light and that's been kind of the big surprises.”

  • Krug likes cable companies for their broadband businesses rather than their video bundles, because broadband is the most robust access route to streaming services, carries pricing power, and earns roughly 70 percent EBITDA margins, while pay-TV margins are materially lower due to content costs.

    “we believe they're close to 70% EBITDA margins, which are very high. And the cable pay TVs margins are actually materially lower”

0.40

Cash flow coverage and downside avoidance are the core defense against permanent credit impairment.

6 pointscentrality 3/5
  • High yield investing is fundamentally about finding what can go wrong, and the firm's main risk focus is issuance quality and cycle progression rather than any single company.

    “In terms of high yield, we're always looking for what can go wrong”

  • Cash flow sufficient to service debt is the key risk reducer; Krug buckets holdings into core, spread-tightening, and opportunistic categories, with core holdings like charter cable broadband benefiting from strong pricing power and roughly 70 percent EBITDA margins on high-speed data.

    “if you've got the cashflow to service the debt, that reduces our risk.”

  • High yield and leveraged loans sit between investment grade bonds and equities: they carry more economic sensitivity than investment grade but earn incremental spread, and they show much less volatility than equities because aggregate high yield leverage of about 4.25x debt to EBITDA is far lower than the roughly 12x EBITDA multiple implied for S&P midcap equities.

    “credit does not experience nearly the same amount of volatility as equities do. If you look at the high yield market in aggregate leverage... is around four and a quarter times”

  • Investors who stretch for yield often overpay, so disciplined credit managers protect enterprise value coverage to avoid permanent impairment, accepting that credit has an asymmetric payoff where bonds are capped near par on the upside and can lose heavily on the downside.

    “we don't want to sacrifice enterprise value coverage, when we try to avoid permanent impairment. If you look at what credit investing is doing, there's asymmetric risk profile, bonds where should a par”

  • Risky issuance is a precursor to future defaults, so monitoring the quality and terms of new high yield and leveraged buyout issuance is the key indicator of where the credit cycle is headed.

    “if typically you have risky issuance, and risky issuance is kind of a precursor to future defaults”

  • Higher-quality issuers typically have the loosest covenants because they have more financing options and lenders accept less protection, so Artisan never buys based on covenants but covenant weakness can cause it to avoid a credit.

    “Typically the higher quality companies have the loosest covenant, which makes sense because they've got availability”

0.36

Misaligned issuer and banker incentives make independence from benchmarks, ratings, and consensus valuable.

4 pointscentrality 3/5
  • Krug views rating agencies as poor judges of credit risk, citing AIG holding a AAA rating months before needing a bailout and AAA structured credit tranches collapsing toward zero.

    “AIG was AAA risk, months before it essentially need to get bailed out and going bankrupt.”

  • Investors learn that incentives are misaligned: issuers want the lowest cost of capital and will tell investors what drives that cost down, and bankers work for issuers rather than investors, so independence is highly valuable.

    “bankers, on the other side are working for the issuer, they're not working for the investor”

  • Companies and bankers are very good at cherry-picking and manipulating data and charts to make conditions look different from a broader sample, which is the most surprising lesson Krug has learned.

    “companies are very good, and bankers are very good at manipulating data. So you actually go and look at the underlying data, they cherry pick.”

  • Krug's portfolios look very different from the benchmark and from peers because the firm values independence and does not care what others perceive.

    “We don't really care what other people perceive, that's why our portfolios look very different than the benchmark”

0.36

Looser loan documents let issuers shift value away from creditors, weakening lender protections.

4 pointscentrality 3/5
  • Benchmarks weighted by total debt give the largest weights to the most indebted issuers, increasing exposure to default risk, which is a structural flaw compared with equity indexes weighted by market cap.

    “having the most amount of debt in being overweight that, can make you more susceptible to default risk”

  • Maintenance covenants have largely disappeared from the loan market, with roughly 85 to 90 percent of issuers now covenant-lite, meaning lenders lost their ability to renegotiate early when profitability deteriorates.

    “It's around 85, 90% issuers now are covenant light.”

  • In downturns, issuers exchange subordinated debt into senior capital at discounts, such as around 50 cents on the dollar, allowing them to cut principal balances; a $1 billion exchange at 50 cents reduces debt by $500 million.

    “Would take subordinate capital and exchange into more senior capital at a discount.”

  • Looser loan documents have let issuers transfer value away from creditors, as seen when Caesars and J.Crew moved collateral out of the creditor group and PetSmart used Chewy collateral to strip value from bondholders.

    “Caesars was an example where they move collateral. J.Crew moved collateral out of the group”

0.36

Lending against enterprise value and business quality protects bondholders better than asset value at par.

4 pointscentrality 3/5
  • Krug's key differentiator is lending against cash flows at par, not asset value, and refusing to sacrifice enterprise value coverage to chase yield, because stretching for yield is expensive and leads to permanent impairment.

    “we don't want to sacrifice enterprise value coverage, when we try to avoid permanent impairment.”

  • Asset value at par did not protect bondholders in the energy downturn: an average EMP bond perceived as high quality fell from par to roughly 35 to 40 cents on the dollar in Q1 2016 as oil dropped from about $100 to $30.

    “the average EMP, as an example, which was perceived to be high quality, dropped roughly 55, 60 points from par to roughly 35 to 40 cents on the dollar, in Q1 of '16. It's that asset value at par basically, didn't help you.”

  • Krug first screens for business quality, seeking high recurring revenue, low capital intensity, higher margins, and strong free cash flow relative to debt, because these traits support enterprise value and downside protection.

    “We tend to get business quality first. So we tend to look for businesses with high recurring revenues, low capital intensity, generally higher margin profiles.”

  • Beacon Roofing's bonds fell about 14 points after a debt-financed acquisition and rate and storm concerns, yielding just under 8 percent to a call two years before maturity, which Krug judged close to par-type risk because repair and remodel demand is resilient.

    “The bonds dropped basically 14 points, and as we look at that, we identified the name through a screen”

0.36

Credit offers defined returns with far less volatility than equities after accounting for default losses.

4 pointscentrality 3/5
  • High yield has had only three negative return years in the last 35 years, and Krug expects forward returns of roughly 6 to 8 percent IRR depending on the portfolio's risk.

    “high yield has had three negative return years than the last 35 years, which is actually very benign”

  • Credit does not experience nearly the same volatility as equities; high yield aggregate leverage is around 4.25 times debt to EBITDA, which Krug says makes the valuation risk far smaller than in equities, excluding periods like 2008 when leverage was loose.

    “Relative to equities, credit does not experience nearly the same amount of volatility as equities do.”

  • The credit opportunity set is easier to model than equities because a bond's return profile is mathematical with fixed upside, whereas equity multiples can move with investor psychology.

    “the opportunity set of the return profile, is much easier to model because it's mathematical as opposed to a psychological where multiples can do whatever they want”

  • A credit spread is compensation over the Treasury rate that discounts expected future defaults and also includes a liquidity premium; with a 2 percent default rate and 60 percent loss severity, defaults alone cost about 120 basis points per year.

    “spreads are a discounting mechanism for expected future defaults and what you're getting paid for that.”

0.24

Data analytics combined with qualitative judgment and credit-specific analysis gives investors an edge.

5 pointscentrality 2/5
  • Krug uses outside data sources, including credit card data, to track trajectories in cable, TV, retail, and consumer spending on a weekly basis, giving his team a real-time edge that many investors do not use.

    “we'll get credit card data as an example, to essentially be able to get confirmation on trajectories of everything from cable, TV spending, to retail spending”

  • Krug believes in a combination of judgment and data analytics: data-driven research is used to reinforce views and provide an advantage, but it must be paired with qualitative assessment rather than replacing it.

    “I think there's an art to it, as well as a science to. I think you need to have both in my opinion”

  • Credit analysts focus on whether a business is defensible, what its cash flow is, and whether it can delever its balance sheet, whereas equity analysts focus on growth rates and beating street earnings expectations.

    “What's the ability of the company to delever its balance sheet. That is the primary metric that we look at”

  • The highest-quality, longest-duration credits were the most vulnerable to rising rates and sold off most, so Krug rotated higher-quality holdings into floating-rate paper that was immune to that risk.

    “it was the most vulnerable because it was taught, it had the most rate sensitivity.”

  • High yield ETFs underperform because fee differentials are small relative to active funds, transaction costs in fixed income are much higher than in equities, and inefficiencies persist that active managers can exploit.

    “Two is transaction costs are much higher for fixed income investing versus equity, where it's almost free to trade.”

0.22

Rating agencies and sector shifts create mispriced opportunities in energy, insurance brokers, and software.

4 pointscentrality 2/5
  • BBB issuance has exploded over the last five years as companies used low rates for buybacks and debt-financed M&A, and vulnerability is greatest in the long-duration investment grade market (average duration about nine years) versus high yield (three to four years).

    “BBB issuance has absolutely exploded over the last five years.”

  • Krug is overweight insurance brokers and software businesses, citing software retention rates in the low 90s before Salesforce-driven growth, predictable cost structures, and strong enterprise value.

    “insurance brokers. They're software businesses, where they have low 90s retention rate”

  • Krug increased energy exposure from near zero in mid-2014 to the high teens as commodity prices made reinvestment unsustainable, identifying names offering 20-plus percent IRRs, extending into MLPs and into suppliers such as Gardner Denver with about half its business exposed to energy.

    “we went from essentially zero energy waiting in the middle of '14 to high teens energy exposure”

  • Rating agencies overrated diversified energy companies at scale and underappreciated cost structure, so low-cost basin producers such as Permian operators with shorter histories were underrated and some were upgraded when oil fell from $100 to $30.

    “they overrated diversified companies at scale, and they underappreciated cost structure, which if you're in a commodity business cost structure is your primary driver”

0.22

Market structure and call features determine how issuers repay debt and how bondholders are protected.

4 pointscentrality 2/5
  • Historically banks lent to companies at roughly three to four times debt to EBITDA and refused incremental leverage, so a public high yield market emerged in the early 1980s as the junior capital solution.

    “the banks would lend to typically, that just depends on the business between three to four times debt to EBITDA. And then when companies wanted incremental leverage, the banks wouldn't take that risk.”

  • High yield bonds typically have seven to eight year maturities, are non-callable for roughly three to four years, and carry a call premium of four to five points, so early repayment requires the issuer to pay breakage, which benefits bondholders.

    “high yield market is typically seven to eight year maturity on average, it is typically non-call for approximately the first three to four years, and there's a call premium of four to five points”

  • Syndicated loans are senior and typically floating-rate with limited call protection, so issuers can reprice them when demand compresses spreads, much like a homeowner refinancing a mortgage when rates fall.

    “it has a limited call protection, which is important because if there's a lot of demand for loans, the spreads will compress and it's similar to a mortgage”

  • Structured credit and CLOs made it cheaper to lend to companies than banks could, shifting loans from bank-originate-and-hold to syndication, and the loan market grew from nothing to roughly a trillion dollars in about 15 years.

    “that market essentially went from nothing to a trillion dollar market in the last 15 years, give or take.”

0.22

Benign defaults and constructive fundamentals keep spreads tight and signal credit's early-cycle leadership.

4 pointscentrality 2/5
  • Spreads are historically near the tight end of their range because fundamentals are constructive, defaults are benign, and are expected to remain benign.

    “spreads are historically... are closer to the tighter end of the range. And part of it is for good reason because the fundamentals are constructive today”

  • Credit tends to be early cycle and leads equities, because credit investors focus on downside, and in the last downturn credit underperformed well before equities did.

    “credit... tends to be early cycle. It tends to lead equities in general”

  • In the October equity volatility, credit fell only about 1.5 percent, and the equity downside concentrated in high-multiple growth stocks, so Krug judged it not systemic and saw no economic deterioration priced into credit.

    “credit was off marginally by about a percent and a half. So the implication there was, it's not a systemic”

  • The key risk Krug monitors is new issuance, since risky issuance precedes defaults; leveraged buyout activity has been light despite roughly a trillion dollars of raised but undeployed private equity capital, which could imply about $2 trillion of debt if deployed at 30 to 35 percent equity.

    “if typically you have risky issuance, and risky issuance is kind of a precursor to future defaults.”

0.20

Credit stress spreads to equities and the real economy by blocking companies' capital spending and funding.

3 pointscentrality 2/5
  • When credit costs rise, the effects spread to equities and the real economy, because companies cannot fund capital expenditure; after the financial crisis, high yield bonds yielding around 20 percent made capital projects uneconomic.

    “When it came out of the financial crisis and the average high yield bond was yielding 20%, it was uneconomic for any company to incur that”

  • When energy company bonds trade in the 40s or 50s, companies lose access to new capital, cannot grow production, shrink output, lay off staff, and have no capacity for dividends or buybacks.

    “if EMPS, if their bonds are in the 40s or 50s, they have no ability to issue new capital.”

  • Credit is a hybrid between equity and investment grade bonds, with high yield and loans offering incremental spread for greater economic sensitivity, and this spread compensates for the added risk.

    “it's a little bit of a hybrid between the two, and at different points of the cycle, they have different characteristics.”