HCI Group: Q2'14 Results Show Business Model Strains

| About: HCI Group, (HCI)
This article is now exclusive for PRO subscribers.


We see business model strains emerging below the surface of HCI's reported results.

Creative reinsurance structures and stock buybacks help camouflage emerging issues, but underwriting income is declining, driven, in part, by dramatic increases in compensation.

We question how much longer investors will pay an inflated valuation for a subsidy dependent business model with negative operating leverage.

Analysis of The HCI Group's (NYSE:HCI) Q2'14 financial results shows a troubled business lies just below the surface of what appeared to be a "business as usual" quarter. Financial reinsurance structures and share buybacks distort the underlying economics of the business and increase the appearance of profitability. However, our adjusted underwriting model shows a clear deterioration of results on a year-over-year basis.

The table below is HCI's underwriting P&L. We have made one key adjustment: We added back $5.1M of accrued, non-cash, reversible profits associated with reinsurance structures. This gives a realistic view of cash reinsurance costs and allows comparisons with Q2'13.


Source: Company filings and estimates for policies and salaries.

Our model includes an estimate of 152,818 policies at the end of the quarter. (Unfortunately, the company has not disclosed the policy count.)

Sequentially, policy count churn and gross premiums were -3.5% and -2.8%, respectively, reflecting natural post-assumption erosion. On a year-over-year basis, the 9% greater policy count coupled with price increases led to 11% growth in gross premiums earned. However, with higher adjusted reinsurance costs, net premiums grew only 3% between Q2'13 and Q2'14, despite the higher policy count.

Loss expenses increased 5.5% to $18.4M from $17.4M, but the ratio of 20.2% to gross premiums is still incomprehensibly low.

Operational expenses increased $6M or 39% on a year-over-year basis. The increased expenses were reflected in a material decline in pre-tax underwriting income of -$5.4M or -23% to $17.7M from $23.1M in Q2'13.

Most of the expense increase was due to discretionary management items. Increased compensation and interest expenses accounted for 70% of operating expense growth. Compensation expense absorbed 8.6% of gross premiums earned in the quarter, a number that appears quite high compared to competitors. As a point of comparison, United Property (NASDAQ:UIHC) spent 4.5% of gross premiums on G&A expenses in the quarter.

The deterioration in results is quite apparent when viewing the basic underwriting business model, as we show below.

Source: Company filings and estimates.

After adjusting reinsurance costs, the total expense ratio for Q2'14 increased to 80.6% from the reported 75.1%, compared to 71.8% in Q2'13. Pre-tax margins, the natural corollary of the total expense ratio, declined to 19.4% from 28.2% in the prior year.

Dwindling underwriting income on a year-over-year basis was largely masked on HCI's P&L by the reduction in reinsurance premiums due to the accrued profit-sharing; most of the reduction in EPS that would have occurred in the quarter was likewise masked by a stock buyback.

During Q2'14, management repurchased 277,510 shares of stock for an average price of $36.03 per share. The table below compares EPS for Q2'14 and Q2'13 on an as reported and adjusted basis.

Source: Company filings and estimates.

As the table indicates, the reported comparison of $1.39 and $1.40 for Q2'14 and Q2'13, respectively, would have been less favorable at $1.36 for Q2'14 adjusting for the stock buyback.

In addition to boosting EPS, the buyback also has the effect of helping to neutralize the impact of increasing management compensation by reducing dilution of share awards. Thus far in 2014, HCI management has granted themselves 98,720 shares.

Buybacks may serve management, but repurchasing stock at 2.4x book value does not serve shareholders well, as illustrated below.

Source: Company filings and estimates.

The table above adjusts book value per share by reducing total assets and equity by the purchase amount of $17.8M and the share count by the number of shares purchased year-to-date. The result is that management destroyed roughly $1 per share in book value with buybacks.

There was one last item in the quarter that struck us as very peculiar. On the conference call, an analyst asked management to discuss the real estate portfolio. Mr. Patel gave an indefinite response. Yet, the 10-Q reveals that HCI entered a $9.8M agreement to finance the development of and option the purchase of a shopping center.

Why wouldn't management discuss what appears to be the single-most important development in its investment portfolio, particularly when directly queried on the topic? Perhaps it is because many may consider it unwise for an insurance company with significant catastrophic exposure to use its small surplus to fund a foray into commercial real estate lending. Or perhaps it was simply an oversight, like the failure to discuss the policy count at quarter's end.


As management pointed out on the call, it is natural for HCI's takeout model to experience policy shrinkage and premium declines at times. Over the last few years investors seem to be willing to accept periodic declines as long as a year-end assumption or acquisition (and all the state subsidies they include), could be counted on to fortify results. However, the favorable insurance environment in Florida coupled with shrinkage at Citizens implies that subsidies may be harder to come by going forward.

The natural dynamics of the business are a declining top-line and increased expense ratios. Creative reinsurance structures and value-destroying share buybacks provide camouflage, but our adjusted model shows eroding underwriting income driven, in part, by a significant increase in compensation expense. Eventually, even the camouflage will prove ineffective and investors will be left contemplating why they should pay 2.4x book value for a business model that, absent subsidies, exhibits negative operating leverage.

Disclosure: The author is short HCI. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it. The author has no business relationship with any company whose stock is mentioned in this article.