Description

Protective Insurance Corp. (“PTVCA/PTVCB” or the “Company”) is a ~$200mm market cap specialty P&C insurer based in Carmel, Indiana. Formerly known as Baldwin & Lyons (prior to name change in 2018), Protective was founded in 1930. The Company focuses on providing coverage to the trucking industry, a narrow market niche, with smaller books of business in workers’ compensation and public transportation fleets. Some of these lines are now in run-off. Protective trades under 0.6x book value (“BV”), a wide discount relative to peers and historical levels. Profitable insurers tend to trade at a premium to BV. While losses over the past several years have limited investor interest (hence the current discount), I believe the Company is turning the corner towards consistent profitability. As this becomes more clear, the stock should re-rate accordingly. I believe there is substantial upside if things go well, without any heroic assumptions PTVCB could double over the next twelve months. At 1.25x BV, this is a $30 stock. 

After pulling back on unprofitable business the past few years, Protective grew NPW by 11% this quarter, driven by achieving an increased rate of 18% on commercial auto policies. The Company is not exposed to pandemic-related losses and will benefit from the surge in last mile delivery and pickup+delivery. The Company has a strong long-term relationship with FedEx and the independent network comprising FedEx Ground. On an annualized basis, Protective wrote $485mm in net premiums this past quarter. 

Typically when insurers trade at such a large discount to book value, it is an indication that the market either believes loss reserves are inadequate or the investment portfolio is mis-marked. Neither is the case here with a vanilla investment portfolio and a 75% adverse development cover on the bulk of reserves which limits further losses. Of critical importance, reserves for each treaty year 2013-2018 have been booked at or above the aggregate stop-loss level. 

The risk-reward skews favorably because the business has been de-risked, limiting further downside while still preserving upside should underwriting results continue to improve on the current trajectory. Even without significant near-term improvement in reported results, investors can still generate outsized returns if the Company is acquired. In June, the board of directors announced a Special Committee to explore strategic alternatives. 

M&A activity in the space has come to life in the past few months with several assets previously viewed as unattractive generating interest from multiple parties. Protective has been on the sale block for a few years with a few false starts, so the stock price does not reflect the potential upside from an acquisition bid. 

Outside of a sale, there are other avenues management and the BOD could pursue to catalyze value here. One is a more aggressive share buyback program, another less likely approach would be a renewal rights deal. 

In the past, I think there was too large of a spread in bid/ask for a deal to occur. In the event that underwriting woes continue, I believe it is reasonable to assume that there is some downside protection from M&A. 

This investment does provide a ~3% current return via the $0.40 annualized dividend. Risk that this is “dead money” for some time in today’s market which has shunned companies like Protective. Index/ETF ownership is low/non-existent. If the management team were more promotional, they might promote Protective as a growth play on the boom in e-commerce.  

Protective is closely held and relatively thinly traded. There are two classes of stock. “A” shares carry voting rights while “B” shares have no votes. PTVCB shares have ADTV of ~11k over the past month while PTVCA barely trades. The classes are otherwise entitled to the same economics. This writeup focuses on the more liquid “B” shares. 

Another risk is that the two share classes are not treated equally. However, the tender for a portion of the “A” shares in May was ultimately rejected. I think the BOD understands the dynamic and am betting they will treat all shareholders fairly. Hereafter, I will use PTVCB interchangeably with Protective.    

Long/short insurance looks like a fertile hunting ground these days. Wide dispersion in valuation for companies that fundamentally are not as different as P/BV (or P/E) multiples would suggest. I think the bigger dispersion relates to management conservatism and stock promotion. Some insurance executives shoot it straight, layer conservatism deep into their corporate culture and thus financial statements and present those financials in an impartial manner. Others promote false sources of competitive advantage (tech capabilities), wax nonstop about growth opportunities and TAM then place pressure across the organization to achieve unrealistic goals. Usually this camp presents the financials in the most flattering light possible. In the short term, the latter approach can work (and has worked), but eventually in insurance, the chickens come home to roost. There is a compounding effect to the factors listed above which adds to the potential reward on both the long and short side. Solid insurance companies conservatively peg loss ratios/reserves and are slow to respond to good news. The opposite is often true for inferior insurance companies. In my view, there are a handful of inferior insurers that have been rewarded insane multiples on inflated numbers. Meanwhile, there are many attractive longs in the space seemingly punished for their modesty and conservatism. 

Commercial auto has been one of the most difficult lines of business over the past 5+ years and Protective has not been spared. The problem has been higher severity driven by an increasingly troubling litigation environment. Several factors are at play, including expanded funding options for litigants, an increasing willingness to litigate and draw out the process and more generous verdicts. 

Looking at annual performance since 2015, a clear trend emerges. After growing aggressively in 2017 and 2018 PTVCB has pulled back significantly. The chart below shows key financial metrics since 2015. Operating EPS declined from $1.60 in 2015 to -$0.19 in 2019 as the combined ratio has remained above 100% since 2017. 

Book value per share has also trended poorly the past few years. BVPS as of September 30, 2020 was $24.18, a decrease of $1.33 per share during the first nine months of 2020, after the payment of cash dividends to shareholders totaling $0.30 per share. BVPS was adversely impacted by investment losses of $10.5 million ($8.3 million after tax, or $0.58/share), the impacts of the updated current expected credit loss (CECL) estimate of $17.0 million ($13.4 million after tax, or $0.95/share) for the MVP Staffing litigation (discussed below) and a deferred tax asset valuation allowance of $1.5 million ($0.11/share). 

The biggest risk PTVCB writes is commercial auto liability. So called “nuclear verdicts” have been, and will likely continue to be, troublesome. It does not seem likely that social inflation will abate any time soon. 

A critical element to this thesis is PTVCB’s aggregate unlimited stop-loss provision on its legacy commercial auto book. This means that for every $100 of additional loss, PTVCB is only responsible for its $25 retention. For the 2020 treaty year (ending July 3, 2020), PTVCB has also placed a reinsurance cover where it retains only 65% of the loss after attachment. The higher retention reflects: (1) a decreased need for stop-loss reinsurance protection resulting from a significant decrease in commercial automobile subject limits profile; (2) a higher cost for this coverage and; (3) PTVCB’s confidence in profitability improvements given the limit reductions and rate increases on commercial automobile products. After July 3, 2020, PTVCB has decided to not renew the reinsurance treaty. Management explained this was due to continued rate achievement in commercial automobile, significant improvements in mix of business and reductions to its limits profile. 

Since the commercial auto market has been experiencing difficulty for a number of years, the existing book of business should receive some benefit from the heady rate increases which have followed and hopefully exceed the elevated loss cost trend. 

Underwriting History:

Historically Protective (fka Baldwin & Lyons) has a reputation as solid underwriters of risk. Commercial Auto has been a tough business for just about everyone who writes it, even the most conservative management teams, so they should be given some leeway but the results have been disappointing nonetheless. The foray into new business lines doesn’t help the cause.   

The pandemic provided some benefit to underwriting from lower loss frequency but will also negatively impact premiums which in many cases are based on exposure. Management does not expect the modest benefits on loss frequency to persist and the current loss picks do not reflect any continued benefit. Likewise with social inflation. The courts may be backed up, but not many insurance executives are ready to suggest that the trend toward higher severity will abate. 

There are some interesting potential tailwinds for commercial auto from increasing penetration of technology for safety and accident avoidance. Also electronic logging. The Federal Motor Carrier Safety Administration's requirements for electronic logging devices went into effect at the end of 2017, but if a driver already had a recording device, it was grandfathered to the end of 2019. 

Protective seems to be ramping up its IT spend significantly this year and I believe the numerous third party options available will allow them to quickly catch up with a reasonable amount of spend. 

Balance Sheet & Investment Portfolio:

One thing that attracts me here is the simplicity of the balance sheet/investment portfolio. Not very many moving pieces or much volatility. 

As of 3Q20, annualized NPW was equal to 1.27x combined stat surplus of insurance subs. So there is no need for incremental capital to write new business. In 4Q20, PTVCB can dividend $52mm from its insurance subs without approval. 

On a GAAP basis, annualized NPW was equal to 1.41x shareholders’ equity as of 3Q20. 

Debt is a small part of capital structure at $20mm vs. $345mm in GAAP shareholders’ equity. 

PTVCB has steadily reduced risk in their portfolio for the past several years by converting equities to fixed income and shortening duration to 2.7 years. 

Total portfolio is ~$1.0bn w/ leverage to equity at ~2.9x. 

Vast majority of portfolio is fixed income at 94.5% with equities/LPs at 5.5%. 

Credit quality is solid. AA- wtd. avg. rating. 14% in BBB, 8% in cash, 6% below IG or not rated. 44% AA- or better. One area of potential risk is the Commercial Mortgage Loan portfolio, but this is limited to only $11mm. 

In years past, there were some concerns about a related party (New Vernon) affiliated with a large shareholder providing investment advisory for a portion of the portfolio. This relationship has been reduced to a point where it is no longer an issue. 

Management and Board:

Jeremy Johnson, CEO: Joined PTVCB in 2019. Spent most of career at AIG including leading Lexington most recently.  

John Barnett, CFO: Joined PTVCB in 2019. Prior roles at First Acceptance.  

Jay Nichols, Chairman: Joined BOD in 2017. Served as Interim CEO from October 2018 to May 2019. Previously CEO of AXS from 2012-2017 and led RenRe Ventures 2001-2010. 

Some well regarded insurance executives on BOD. 

Throughout most of its history, Protective (Baldwin & Lyons) was a sleepy company with significant long-term ownership by the Shapiro family. Over the past five years, there has been significant management and BOD turnover. 

In May 2016, then-CEO Joseph DeVito, a company veteran since 1981, resigned in protest over leadership changes and strategy issues. Two other long-time execs left at the same time, the Deputy Chairman and former CEO as well as the CFO. This was reportedly due to conflicts with Steven Shapiro, who was appointed Executive Chairman several months prior. Several other execs left around this time or shortly thereafter. 

DeVito’s statement is pasted in part below:

“The elevation of Steve Shapiro to the position of executive chairman and subsequent actions and decisions made by Steve have significantly reduced my authority, and thereby my ability, to properly and appropriately run the company,” he wrote.

“This is not intended to be a criticism of Steve as a person, however, in my opinion, he does not have the necessary background or experience to serve in that capacity. I believe he was assigned that position, not due to any qualifications or background that would merit the appointment, but rather as a result of his family’s ownership of stock.”

At this time, Randall Birchfield was named CEO, Michael Edwards was named Principal Accounting Officer and Michael Case was named COO. In August 2016, Birchfield is also named President and William Vens is named CFO. 

In May 2017, Jay Nichols was appointed to the BOD. In August 2017, Edwards is terminated and Vens also adds Principal Accounting Officer title. 

In Feb. 2018, Steven Bensinger was added to BOD and Birchfield adds COO title. Prior COO (Case) leaves company. 

In May of 2018, Steve Shapiro, who had served since 2007, leaves BOD. 

In August 2018, Birchfield signs a new employment contract running through 2021. Then in October of same year, Birchfield resigns. Birchfield is replaced by Jay Nichols as interim CEO. In November 2018, Protective reports a disappointing quarter with further adverse loss development from claim severity. 

In February 2019, Norton Shapiro, a long-time member of the BOD, died at age 85.

In May of 2019, Jeremy Johnson is named CEO. In June of 2019, William Vens left Protective to join another company and in September John Barnett was appointed CFO. 

Two members of the Shapiro family, Nathan (83) and Robert (81) currently sit on the board. Stephen Gray (66) serves as trustee to the Shapiro family and is not independent. 

This is a lot of turnover. And it should be a red flag. But it seems like things have stabilized here for the time being. I don’t have much unique insight on management or the BOD here, but I have heard some positive things. I like that they appear more conservative than most. Management is compensated based on underwriting income and there are adequate incentives in place to align interests. 

The Shapiro family owns ~41% of Class A (voting shares) and at least 21% of Class B shares. 

M&A Activity:

It is no secret that Protective has been on the block for the past few years. It seems there is a significant bid-ask spread. 

VJ Dowling, a long-time insurance analyst/investor, owns ~6.5% Class A shares. He joined with David Delaney, CEO of Lancer Insurance (a competitor to Protective) earlier this year to tender for 2.6mm Class A shares. The Shapiro family agreed to sell 1mm shares at $18.30 as part of the agreement. PTVCB opposed the tender and Lancer has since left the group, according to an updated filing in August.  

M&A activity in the insurance space is heating up as COVID-related losses are not as bad as feared and more signs emerge of a hardening market.  

Recent take-outs of ALL-NGHC, State Farm-Gainsco, ACGL & ESGR-WTRE, Zurch/Farmers-Met P&C, etc. Starting to see multiple parties compete for some deals. 

There are also several well-funded start-ups that are looking at buying a platform. I have seen some rumors that the former Arch CEO was involved in the bid process at some point.  

Norton Shapiro’s passing and the involvement with the Dowling group on tender create an interesting dynamic. There is risk that the Shapiro family ownership (or individual members of the family) creates an overhang on the relatively illiquid shares in the event of a hasty exit. 

To be clear, I don’t have any special insight into the motivations of the various parties here. I do believe that the current M&A market provides both upside potential and downside protection for PTVCB shareholders.  

Valuation:

Since they have been basically breaking even and thus not earning the cost of capital, you could make an argument that a discount to BV/TBV is warranted. However, this calculus will change if they can turn the corner over the next few quarters on underwriting profitability. There has been sequential improvement and the Company has been getting a lot of rate over the past few years. If management can successfully turn this around, there is a lot of upside even at conservative multiples. The chart below provides some quick+dirty math. 

Management believes they have line of sight to a 10% ROE at a 96% combined ratio. If this can be achieved it would warrant a decent premium to BV. Using current figures, a 10% ROE and 96% combined ratio foots to Operating EPS of $2.30-$2.40 per share. 

If we are past the worst for commercial auto, they will see a significant earnings recovery because of how much rate PTVCB has taken. 

I think PTVCB is an attractive asset to a wide variety of potential acquirers. Also think you have a situation where if things get worse, the lower expectations on sale proceeds will serve as a catalyst to help realize value. 

Capital Management:

The Company has been conservative and methodical with deploying its buyback, only repurchasing $1.8mm worth of stock YTD. Given the BV accretion this could be additive to the story. 

In Feb. 2019, PTVCB cut its dividend from $0.28 per quarter to the current $0.10 per quarter. 

Recent Tender Offer:

From PTVCB 10-Q:

On May 5, 2020, the Board of Directors (the “Board”) of Protective formed a special committee of independent directors (the “Special Committee”) to evaluate a Stockholder Support and Contingent Sale Agreement (the “Contingent Sale Agreement”) entered into by and among certain prospective third party purchasers (the “Offering Parties”), certain of Protective’s shareholders and the other parties thereto. We received notice of the Contingent Sale Agreement on April 23, 2020, the date the Offering Parties filed amendments to Schedule 13Ds relating to our Class A Common Stock.  The Contingent Sale Agreement was amended and restated on August 17, 2020.  Subject to the satisfaction of certain conditions under the Contingent Sale Agreement, the Offering Parties may commence a tender offer to purchase all of the outstanding shares of Protective’s Class A Common Stock.

In June 2020, Protective announced that the Board determined the transactions contemplated by the Contingent Sale Agreement are not in the best interests of Protective and our stakeholders. As part of this evaluation, the Board determined that it would also recommend against the potential tender offer contemplated by the Contingent Sale Agreement if it were commenced, and that if the transactions contemplated by the Contingent Sale Agreement were consummated, it expects to take the necessary actions to redeem all or certain of the Class A shares of Protective purchased by the Offering Parties pursuant to Protective’s Code of By-laws.  Protective also announced that the Special Committee of the Board is exploring, with the assistance of its independent financial and legal advisors, strategic alternatives that may be available to Protective. There can be no assurance that the Special Committee or Board will determine that a strategic alternative is in the best interest of the Company and its stakeholders, or that a transaction will be entered into or, if entered into, the timing, terms or conditions thereof.

During the second and third quarters of 2020, we incurred an aggregate of $2.1 million ($1.7 million, net of tax) of expenses in conjunction with the Board’s review of the Contingent Sale Agreement and the activities of the Special Committee.

Personnel Staffing Group Litigation:

Protective has been engaged for over a year in litigation with a former policyholder, Personnel Staffing Group (“PSG”) also known as MVP Staffing (“MVP”) regarding a workers’ compensation insurance policy. Detailed narrative below from PTVCB 10Q. The dispute has gone back and forth in search of proper venue. Ultimately, management believes the suit is meritless and they expect to collect the full amount due, but there is a sizable allowance on the receivable which was increased by another $1.5mm this quarter. Management reiterated that the incremental CECL credit allowance does not reflect views on ultimate outcome.  

Protective’s max. exposure to loss is roughly $2 per share to BV with ability to gain ~$1 in BV if litigation is resolved (have an allowance currently). PSG/MVP does not seem to have much basis for a case, but an unfavorable outcome appears well discounted at PTVCB’s current valuation. 

As an aside, MVP Staffing received significant stimulus funds from the PPP. The business address and suite number for its Illinois location alone appears to have received $38 million across several different entities. This same company has also been involved in lawsuit for similar actions with prior insurer Zurich. 

From 10-Q: 

In July 2019, Protective Insurance Company (“Protective”) was named as a defendant in an action brought by a former insured, Personnel Staffing Group d/b/a MVP Staffing (“PSG”), in the U.S. District Court for the Central District of California (the “California Action”) alleging that Protective had breached its workers’ compensation insurance policy and had breached the duties of good faith and fair dealing. Protective provided workers’ compensation insurance to PSG from January 1, 2017 through June 30, 2018, which was subject to a $500 per claim deductible to be paid by PSG.  No specific damages were included in the complaint.  In August 2019, Protective filed a motion to dismiss or stay the action.

In August 2019, Protective filed a lawsuit against PSG in Marion County Superior Court, in Indianapolis, Indiana (the “Indiana Court”) alleging breach of contract, breach of the parties' collateral agreement, breach of the parties' indemnity agreement, and seeking a declaratory judgment regarding PSG’s obligation to fund its ongoing claim deductible obligations and adequately collateralize Protective’s current and ongoing claims exposure pursuant to terms of the parties' agreements (the “Indiana Action”).  In October 2019, Protective amended the complaint to include allegations of misrepresentation as to source of coverage, negligent misrepresentation, fraud and racketeering and seeking injunctive relief.  In November 2019, PSG filed a motion to dismiss the Indiana Action on the basis of comity with the California Action, claiming that California was the proper forum for Protective’s claims.

In February 2020, the Indiana Court issued an order dismissing the Indiana Action without prejudice; the Indiana Court declined to rule on the legal effect of the forum selection clause in the parties’ agreements, finding that any interpretation should be addressed by the court in the California Action.   On April 28, 2020, Protective’s motion to dismiss the California Action was granted without prejudice on grounds that Indiana is the more appropriate forum.  On May 4, 2020, PSG filed a notice of appeal in the 9th Circuit Court of Appeals, challenging the order of dismissal in the California Action.  On May 1, 2020, Protective filed a motion with the Indiana Court to re-open the Indiana Action, which was denied on September 23, 2020, pending resolution of PSG's 9th Circuit appeal.  Protective intends to vigorously pursue its claims against PSG, however, the ultimate outcome cannot be presently determined.

Pursuant to the terms of the workers’ compensation policies, Protective has a duty to adjust and pay claims arising under the policies regardless of whether PSG makes payments to Protective for deductible obligations under the policies.  Under its contractual obligations to Protective, PSG is required to maintain a “loss fund” for the payment of claims, the balance of which is to remain at or above $4,000; in addition, PSG is required to provide collateral in an amount equal to 110% of Protective’s current open case reserves on workers’ compensation claims arising under the policies.

As of September 30, 2020, Protective had approximately $19,400 in receivables on claims arising under PSG’s workers’ compensation policies and had exhausted all collateral provided by PSG.  Protective continues to pay claims settlements under the policies without reimbursement from PSG.  For the past six months, the average monthly invoices have been approximately $786.  PSG’s estimated ultimate obligation under the agreements is approximately $46,730 as of September 30, 2020 (inclusive of the $19,400 in receivables noted above).  At September 30, 2020, based on the Company's assessment that PSG will continue to operate as a business and that the terms of the agreement with PSG will be legally enforceable, the Company believes that it will fully collect all current and future amounts due from PSG relating to this matter.

The Company included this matter in its assessment of the impact of adopting ASU 2016-13, the new guidance for measuring CECL, which is discussed in Note 1.  A probability-of-default methodology was applied to projected estimated cash flows to estimate the allowance for expected credit losses for this matter.  The Company considered the delay in reimbursement for claims paid as well as probability of default assumptions when analyzing the credit loss related to this matter.  As of January 1, 2020, in conjunction with the adoption of ASU 2016-13, the Company recorded an allowance for expected credit losses of $15,000 ($11,850, net of tax) as a reduction to equity.  During the third quarter of 2020, the Company performed an update to its CECL allowance calculation related to the PSG matter.  As noted above, there have been further delays in the litigation process, which have extended the estimated cash flow timing.  As a result of these delays and an increase in the estimated ultimate obligation, the Company recorded an additional allowance of $1,500 ($1,185, net of tax) within other operating expenses in the condensed consolidated statement of operations for the three and nine months ended September 30, 2020.  In the event of a situation that results in no recovery from PSG, the Company would incur an estimated charge to the condensed consolidated statement of operations of $30,230 ($23,882, net of tax), which represents the estimated ultimate obligation discussed above less the CECL allowance.

Risks:

There are a number of things that could derail this thesis which are discussed below, but I think there is enough margin of safety between valuation (@ 60% of BV), adverse development cover and the actions management has taken to reduce risk and capture rate in excess of loss trend.  

Adverse Development/Continued Problems w/ Severity: This would represent a continuation of the existing trend. Although most prior year losses are covered up to 75%, investors would not view this favorably and I believe even if recent accident year loss picks are left unchanged, the market would severely discount the earnings stream. This is probably the most significant risk Protective faces.      

Ratings Downgrade from AM Best: Management has expressed a commitment to maintain current ratings and its actions jive with this commitment as PTVCB has been conservative on the capital management front. A ratings downgrade would serve as a catalyst for a sale, but definitely lowers the potential value. I think this risk is adequately discounted.

Illiquid Stock: This idea is a small/micro cap and relatively illiquid with a high degree of insider ownership which could cause dramatic price action. 

Macro Economic Weakness/Capital Markets Sober Up: Protective’s business is not as levered to the wider economy as most, although they would lose premiums from a reduction in miles driven/employment. The Company’s investment portfolio is conservative. If capital markets cool down, this would be a negative for a potential M&A catalyst. Would also likely hurt the relative valuation where PTVCB screens very well right now. 

Catalyst

Catalysts include:

-4Q20 and 1Q21 earnings

-M&A

-Share repurchase activity