Admittedly, the title to this article may be a pun too far. However, in this post, we’ll show you where you could have found signs of potential distress within rental car company Hertz Global Holdings Inc. (HTZ.N), well before it filed for Chapter 11 bankruptcy protection on May 26.
The analytics presented in this report come from Refinitiv’s suite of StarMine quant models. As shown in Exhibit 1, the Eikon widget that sorts StarMine models into bullish and bearish categories show many bearish indicators for Hertz and not a single bullish one.
StarMine models fall into two categories – alpha generating stock ranking models and credit default risk. Both are ranked on a 1-100 percentile basis, with 100 being best. Note for Hertz, all four credit risk models receive a score of 1.
Having much of your revenue-generating assets parked at airports and a balance sheet bloated with debt is not a place a company wants to find itself with people isolating at home during a global pandemic.
Exhibit 1: StarMine model scores for Hertz
Source: Eikon by Refinitiv, StarMine
The most comprehensive and predictive of the StarMine credit risk models is Combined Credit Risk. As shown in Exhibit 2, it was calculating a Probability of Default of nearly 52% over the next 12 months when Hertz filed for bankruptcy protection. That maps to a Model Implied Rating of CC and a rank of 1 out of 100. The Model Implied Rating is a mapping of what a rating agency would likely assign to a company with a comparable probability of default.
Among U.S. companies with market caps of greater than $100 million, only one company has a higher probability of default, Chesapeake Energy. This is shown in Exhibit 17 at the end of this article.
Exhibit 2: StarMine Combined Credit Risk Model Summary for Hertz
Source: Eikon by Refinitiv, StarMine
As the two-year model history chart shows in Exhibit 3, a low Model Implied Rating wasn’t just a recent phenomenon – Hertz sat at CCC+ on May 28, 2018. This was well below the actual rating agency’s B+ rating. It remained below the agency’s rating throughout these two years.
Refinitiv research has shown that when this Model Implied Rating is different than a rating agency’s, the agency is four to five times more likely to revise in the direction of the Implied Rating than in the other direction.
Exhibit 3: Combined Credit Risk Model Implied Rating History Chart
Source: Eikon by Refinitiv, StarMine
That’s exactly what happened in the case of Hertz. S&P made the first of several downgrades, to B-, on April 20, 2020. By that date, the Model Implied Rating had declined further, to CC.
Exhibit 4: A close-up on recent rating activity
Source: Eikon by Refinitiv, StarMine
The three input components to the Combined Credit Risk model are shown in Exhibit 5. They are the Structural Credit Risk, SmartRatios Credit Risk and the Text Mining Credit Risk models. Each takes a different approach to the same issue – trying to calculate a probability of default (which would include bankruptcy).
We’ll next examine the SmartRatios model, which, like the Combined Model, also ranked Hertz as 1 and with a Model Implied Rating of CC.
Exhibit 5: Combined Credit Risk model input components
Source: Eikon by Refinitiv, StarMine
Exhibit 6: SmartRatios Credit Risk Model Summary
Source: Eikon by Refinitiv, StarMine
Also like the Combined Model, the SmartRatios Credit Risk (SRCR) model had an implied rating well below that of the actual rating agency for at least year prior to bankruptcy.
Exhibit 7: SRCR Model Implied Rating History
Source: Eikon by Refinitiv, StarMine
Exhibits 8 and 9 show the input details of this model. Each model component gets its own 1-100 relative rank. Low scores detract from the overall score. Those near the bottom are indicated in red. Much like a human analyst, the model looks at a number of ratios over the categories of profitability, leverage, coverage, liquidity, and the growth and stability of earnings, revenue and returns.
Profitability is negative, over both the trailing twelve month (TTM) actuals and next twelve month (NTM) forecasts. Forecasts, when available, are based on the StarMine SmartEstimate®, which reweights the consensus to place more weight on the more recent estimates and more accurate analysts.
The high degree of leverage on Hertz’s balance sheet is indicated by a Net Debt/Equity ratio of greater than 10.
Exhibit 8: SRCR Model Component Details
Source: Eikon by Refinitiv, StarMine
Under the category of Coverage, Hertz came into this economic crisis with low interest expense coverage and negative free cash flow. Liquidity calculations showed little surplus cash on the balance sheet. Return on Equity (ROE) growth was negative and earnings were highly volatile. All this contributed to the model’s overall score of 1.
Exhibit 9: SRCR Model Component Details, Continued
Source: Eikon by Refinitiv, StarMine
Turning from credit risk models to a financial statement-based alpha-generating earnings quality model shows further red flags in Hertz’s financial fundamentals. The StarMine Earnings Quality (EQ) model ranks companies based on its calculation of the sustainability of earnings. It has academic underpinnings in what has become known as The Accruals Anomaly. Hertz receives a score of 5 out of a possible 100.
Exhibit 10: StarMine Earnings Quality Model Summary
Source: Eikon by Refinitiv, StarMine
Exhibit 11 shows the details driving this model score. Of the four components: Accruals, Cash Flow, Operating Efficiency and Exclusions, the low scores are in Cash Flow and Operating Efficiency.
Exhibit 11: StarMine EQ Model Component Details
Source: Eikon by Refinitiv, StarMine
The cash flow component examines a company’s free cash flow generation. It is a basic measure of cash flow from operations (CFFO) minus capital expenditures (CapEx). The issue here can be visualized in Exhibit 12. Comparing quarterly capex to CFFO, the bar chart shows that in every quarter over the five years shown, capex greatly exceeds CFFO.
Exhibit 12: Hertz CapEx vs. Cash Flow from Operations
Source: Eikon by Refinitiv, StarMine
The result is the consistently large negative free cash flows seen in Exhibit 13. In this bar chart, the top of the red bar represents GAAP net income, the bottom of the bar is free cash flow. In the most recent quarter, ending March 31, 2020, net income was a negative $356 million. Free cash flow was a negative $3.956 billion.
On a GAAP basis, Hertz has lost money or barely broken even during most quarters over the last five years. In this example, a white portion of these bars indicates the amount by which net income was below the $0 line.
Exhibit 13: Hertz Free Cash Flow vs. Net Income
Source: Eikon by Refinitiv, StarMine
Exhibit 14 is a time series representation of the StarMine EQ model’s measure of operating efficiency. This model component is a sector relative rank of a company’s average return on net operating assets. It is a combination of the company’s pretax profit margin and its asset turnover.
This view is comparing the return ratio for Hertz in blue to the industry median return in gold. In the most recent quarter, Hertz generated a return of only 2.5% vs. the industry return of 13.3%. Except for a brief recovery in 2018, Hertz has seen a decline in its returns. At these levels, the company’s return on capital most surely was below its cost of capital – a sign of shareholder value destruction.
Exhibit 14: Return on Net Operating Assets
Source: Eikon by Refinitiv, StarMine
The EQ model provides additional information about a company’s financial fundamentals in the auto-text generated bullet points in Exhibit 15. These notes were updated when the model first processed the company’s latest Form 10-Q filing.
Among other possible warning signs, notice that the debt-to-equity ratio increased, even with a $2.235 billion sale of assets.
Exhibit 15: StarMine EQ Financial Highlights
Source: Eikon by Refinitiv, StarMine
An alternative data source provided yet another early warning signal that not all was well with Hertz. In Exhibit 16, we join data from third-party alternative data provider, Minneapolis-based LinkUp, whose job market data is now on our business partner’s BattleFin Ensemble Platform, with our own I/B/E/S Estimates data and Refinitiv’s StarMine SmartEstimates data.
The green line and left vertical axis show the number of Unique Active Jobs for Hertz that LinkUp compiled from job posting sites. That number dropped precipitously from around 2,500 to near zero between mid-March and the end of the month.
Similar to a trend we noted for Applied Materials, taking down job postings preceded sell-side analyst estimate cuts: https://refini.tv/2VSFKEQ
Led by the StarMine SmartEstimate, for the quarter ending in June, both it and the I/B/E/S consensus revenue estimate didn’t see aggressive cuts begin until a week or so after the job boards were cleared.
Exhibit 16: Hertz Unique Active Job Postings data from LinkUp vs. June Quarterly Revenue I/B/E/S and StarMine Estimates Data from Refinitiv
Sources: StarMine and I/B/E/S data from Refinitiv, LinkUp data, available on BattleFin Ensemble Platform
You can use the Screener App within Eikon to rank any investable universe, index constituents, or geography by our various credit risk models, probability of defaults or Earnings Quality model scores.
Topping the list when sorting by Probability of Default in our Combined Credit Risk model, Chesapeake Energy Corp. has reported to be on the brink of a Chapter 11 filing itself. Hertz is in spot number two. https://www.reuters.com/article/us-chesapeake-enrgy-bankruptcy-exclusive/exclusive-chesapeake-energy-preparing-bankruptcy-filing-idUSKBN22B31M
But, taking our lead from Audit Integrity, we’ll leave it to you to fill out the remaining 18 of a “top 20” list.
Exhibit 17: Screening the StarMine Combined Credit Risk Model’s Calculated Probability of Default Percent in Descending Order
Source: Eikon by Refinitiv, StarMine
