Description

Cheap hotel business with great capital allocation, substantial pockets of upside, good balance sheet and shareholder friendly management. We believe the shares are worth more than double.

All $ values are in CAD.

Summary

Holloway Lodging Corp (“HLC”) is the owner of primarily limited service hotels in Canada, spanning 34 properties and 3,854 rooms. HLC’s common stock is undervalued on a relative and absolute basis. The company has a solid balance sheet, generates substantial free cash flow, has opportunities for growth through acquisitions and hotel renovations, and will benefit from continued shareholder-friendly capital allocation by the controlling shareholder and manager.

HLC is the least expensive hotel business in North America across a number of metrics. It trades at an implied cap rate of 12.0% compared with a median of 7.8% across comps, and a price to free cash flow of ~6x compared with a comp median price to FFO of 9.5x (FFO is a more aggressive measure, since it adds back D&A and does not deduct capital expenditures).

The company is controlled by Clarke Inc., a Canadian activist investor. Clarke has historically monetized its large holdings and redeployed its capital.

Given the undervaluation of HLC, the safety of the business and balance sheet, and aligned controlling shareholders and management team, we believe HLC is an attractive stock that has the potential to earn IRRs north of 20% over a multi-year holding period.

Brief history

Clarke was a shareholder, and had board control, of Canadian hotel company Royal Host. In July 2014, it engineered the merger with Holloway, with which it was involved since 2010. Both businesses had been very mismanaged and overleveraged before Clarke’s arrival. There is a good introduction to Clarke and its CEO Michael Rapps in this video.

Capital Structure

HLC has 19,163,666 shares of common stock outstanding. It has two series of convertible debt totaling $93.1m in face value (both would require a share price of $25 to make conversion attractive, so dilution is not an issue). The debentures cost 6.35% for the Series B and 7.5% for the Series C. The company also has a $8.5m credit facility at prime plus 1.5%. Finally, it has various mortgages totaling $121m at 6.2%.

Total debt is $235.8m with a WA cost of 6.4%, which can be lowered, albeit slowly. The mortgages have defeasance provisions which prevent early refinancing, and approximately 70m of the mortgages sit inside CMBSs which won’t mature until 2017.

HLC does have a number of properties that are unencumbered (the company has nineteen mortgages and 34 hotels), which could provide cheap capital for acquisitions or for buying back the debentures, which frequently trade below par.

Valuation and FCF calculation

To calculate HLC’s free cash flow we have taken reported numbers for the TTM, since this already includes the impact of the oil and gas decline (although perhaps there is more pain to come – this point is not certain).

HLC’s business is seasonal, with Q1 being the weakest quarter. Q2 and Q4 are typically at similar levels, and Q3 is the strongest quarter. Adding up the last four quarters we arrive at FCF of $9.4m.

Here is how we think of free cash flow. First, we take reported net income and add back non-recurring items and deduct non-cash gains, reversals and one-time gains. We also deduct/add back any income tax provisions and recoveries that are non-cash in nature. This results in adjusted net income.

Then, we add back reported D&A, and deduct maintenance capex. The hotel industry guides to 4% of revenues for maintenance capex, and this is what HLC uses in its capex reserves. We however choose to use 6% because we believe this is a more realistic number. Furthermore, HLC uses 4% over revenues excluding food and beverage, while we use 6% including food and beverage. Again, this is more conservative. We’ve found that since hotels need to renovate every 7-10 years, the 4% figure is unrealistic and leads to an inflated estimate of free cash flow.

Our resulting FCF number of $9.4m means the shares trade at ~10x free cash flow. Note that if we went back to Q3 2014 the TTM FCF number would have been $12.6m; at the end of 2014 the TTM number was $13.3m; at the end of Q2 2015 the TTM number was $10.7m.

It is difficult to pinpoint the reasons for the decline but there are two large hotels which have been under renovation (and have recently re-opened). In 2014 there was also a flood/fire in Northern Ontario, and people were airlifted to Thunder Bay, resulting in a very good 2014 for HLC’s property there.

But there are a few additional adjustments we must make to our $9.4m current TTM estimate.

First, we must add back $1.1m in NOI resulting from the hotel renovations in Q3 2015 (our source for this estimate is the top of page 5, Q3 2015 MD&A). Then, we must deduct $518k attributable to the sale of the Holiday Inn Express in Myrtle Beach, SC in December 2015.

Then, we add $732k which is the result of lower debt expense resulting from the sale of the aforementioned SC hotel and the pay down of the credit facility with cash left over from that sale. We assume that a $5.3m loan receivable by HLC will be used to pay down mortgages or buy back debentures (this $5.3m loan is expected to be repaid in H1 2016 and is the result of the sale of the company’s franchise business).

All of this gets us to ~$10.7m in FCF. Now, in Q3, Rapps mentioned on the call that once reopened the two renovated hotels should produce NOI levels 2-3x prior to the renovation.

We assume the bump is 2x. Given that the $1.1m decline occurred in Q3, which is typically 40% stronger than the average of the other three quarters, we estimate that the NOI for the other three quarters is around $788k per quarter. Therefore, the total bump in NOI we should expect is $1.1m + 3*$788k = $3.5m.

Finally, our last adjustment results from the distressed Days Inn in Sydney, which was acquired through its mortgage, foreclosed on, renovated and should now be producing $0.5m in NOI. (This was a home run deal by Rapps, as the all in cost was only $3.3m, resulting in a 15% unlevered return on total cost.)

Our total FCF estimate therefore reaches $10.7m + $3.5m + $0.5m = $14.65m. That’s $0.76 per share, putting the shares at ~6x or a 16% free cash flow yield. (Note that our calculation isn’t exact because we are not increasing the maintenance capex associated with the additional hotel revenues, but it is “close enough” for our purposes. Our assumptions on the seasonality of the closed hotels being similar to that of the overall portfolio could also be off.)

On the enterprise value calculation, we adjust current debt of $235m down to $222.5m to take into account the divestitures and debt pay downs envisaged above. Enterprise value is therefore market cap of $90m plus $222.5m or $312.6m.

The NOI calculation follows a similar bridge; we take the TTM NOI figures and adjust upwards by $1.1m for the lost NOI due to the renovations; downwards by $518k due to the sale of the SC hotel; upwards by $3.465m for the incremental NOI once the newly reopened hotels stabilize, and upwards by $0.5m for the Sydney Days Inn, for a total TTM NOI of $37.5m.

NOI/EV is HLC’s cap rate, 37.5/222.5 = 12.0%. Cap rate is a metric used to value real estate. This is pretty high, but not far from one off transactions HLC has been able to complete. The company acquired three hotels in 2014-2015 at 11-11.6% cap rates. It also sold four properties with cap rates ranging from 2.6% to 7.6% (the very low cap rate was for the Travelodge in Toronto, which generated a minimal amount of cash flow).

Stylized 20 percent IRR in a static scenario

Below is an exercise showing how a 20% IRR can be delivered simply from free cash flow accretion shifting value from debt to equity. Note that the NOI is static, which is unlikely (it’s more likely to grow).

Where do we get the 9.5% exit cap rate? La Quinta currently trades at a 9.8% cap rate and some analysts have noted it appears cheap, with the shares down 50% over the last 12 months. That is possibly a decent comp for HLC. Furthermore, if we assume 40% of HLC’s NOI is from western hotels (which are perhaps disproportionately profitable due to the oil and gas industry exposure), and that the balance is from non-oil and gas hotels, and assign market cap rates to each, we arrive at a blended cap rate of 9.5%. Again, this is probably roughly right, and hard to prove, but we take a stab at it:

Finally, we’ll note that the entry free cash flow yield of 16% would represent an IRR of 16% assuming no cap rate compression (multiple expansion) and no growth or decline.

Renovation upside

HLC’s two largest hotels have been renovated recently and reopened in January. The Hilton London was renovated and rebranded DoubleTree by Hilton (323 rooms, HLC’s largest hotel). The Chimo Hotel was renovated and rebranded Holiday Inn Ottawa East (256 rooms, HLC’s second largest hotel).

The renovations cost around $20m split evenly. Michael Rapps said in the Q3 conference call that he expects the NOI to be 2-3x the level before the renovation. We’ve backed into this above and believe the incremental NOI at 2x prior level to be around $3.465m. Assuming the market cap rate for these hotels to be around 8.5% (per management), the incremental value created on the $20m renovation is $3.465m/8.5% = $40.76m. This is very meaningful at 13% of EV and 43% of market cap, and an excellent return on unlevered capital.

Furthermore, if the total new run-rate NOI for the hotels at 2x is $6.9m, they would be worth $81.5m, a whopping 26% of EV and 88% of market cap. A sale of these hotels would therefore be extremely accretive to HLC.

Here’s what would happen to our blended cap rate if these hotels were sold:

The appropriate blended cap rate would shift from 9.5% to 9.7% but the reduction in EV would more than compensate for it:

One of the interesting pieces of the renovation is that the London Ontario, being a Hilton, had a pretty high cost structure, because the hotel must be operated to Hilton standards. However, it can’t command the ADR necessary to make that cost structure worthwhile. With the conversion to a Doubletree, the cost structure will be lower and more in line with achievable rates.

The Chimo Hotel renovation has already driven a substantial increase in ADR for groups. Management gives a good narrative of this and more in its Q3 conference call.

Development upside

The Chimo Hotel (now Holiday Inn Ottawa East) is not only advantageously located across the street from Ottawa’s largest shopping center (see map), it also has expansion land for a structured parking lot and a substantial amount of real estate. According to management, the city would like to see density, as in ~20 stories. Potential uses could include condos, rentals and offices. Demand is not strong at the moment but is expected to improve as the city completes its light rail transit over the next few years.

In addition, the St Laurent Shopping Centre has filed papers for an expansion (currently it has under 1m sf of GLA), and several government buildings are planned for a site half a mile away.

Across town the Travelodge Ottawa West (see map) is another potential development site. Here there could be a large residential component.

It is difficult to estimate the potential upside from these possible developments but if they happen they would be icing on the cake.

Distribution of hotels across Canada

You view here an image and an interactive Google map link of HLC’s hotels. There is more information on the company’s website.

Pockets of hidden value

NOLs

Holloway has substantial NOLs. When asked, the company said that to put it bluntly, it won’t pay cash taxes for a very long time, at least ten years. We are not modeling taxes in our free cash flow calculation.

The magnitude of the NOLs is difficult to ascertain because it’s not disclosed properly. The disclosure in the MD&A indicates tax assets of $22.7 million but the “real” number may be twice that. Furthermore, there are UCC tax credits of around $330m. The way these get used up is the following: if the company spends $20m renovating its hotels, the UCC tax credit is reduced to $310m. It is unclear the mechanics of this but it is a further tax shield the company can use.

Land adjacent to hotels

HLC has substantial excess land at some of its hotel sites. This land could either be monetized or developed for further upside (see below).

CMBS debt

Several of HLC’s mortgages are in CBMS structures and cannot be readily refinanced. However, as these structures mature over the next could of years, HLC could materially lower its cost of debt. At 6.4%, the cost of debt is too high. Each 50bps in debt reduction corresponds to roughly half a turn in free cash flow multiple, so debt cost reduction would be substantially accretive to HLC. We note that last year the company refinanced one of its mortgages, lowering the cost from 6.0% to 4.25%, a very meaningful reduction.

Reasons for undervaluation

There are several potential factors which could explain the observed undervaluation:

• Limited liquidity: the average daily amount traded is a mere $42k

• Practically no analyst coverage: there only appears to be one Canadian sell-side analyst following the stock

• Small issue and float: market cap of $90m and free float of around half of that limits supply of tradeable shares

• Oil and gas exposure: about one third of HLC’s hotel rooms are in Western Canada and exposed to the oil and gas producing parts of the country

• Hotel cycle: after years of RevPAR growth, investors are wary the cycle is turning

• Airbnb threat: unlocking underutilized hotel rooms has the potential to curtail any future RevPAR growth and disrupt the economics of the hotel industry

Factors mitigating undervaluation

• Airbnb threat is likely not a concern in the smaller markets served by HLC’s limited service hotels

• According to research firm HVS, the value of most of Canada’s hotels is expected to continue rising in the coming year, owing to good supply/demand dynamics – link to report

• Oil dynamics have been playing out since mid-2014 and are likely priced into the shares

• There are very few publicly traded hotel comps in Canada, however HLC is the cheapest among them and among the much larger comp set in the U.S.

• HLC’s 12% implied cap rate is even cheaper when considering the fact that interest rates in Canada are lower than in the U.S. This image shows how the Canadian sovereign yield curve was very similar to that of the U.S. in Jan 2005; today, nearly the entire Canadian sovereign yield curve fits inside that of the U.S., trading around 60-80 bps lower – link to image

• The company has considerable “self-help” upside including lowering its debt costs, pursuing accretive acquisitions, hotel renovations, and can deliver IRRs in the 20s even in a static scenario

• Absent compelling opportunities to deploy capital in hotel renovations and acquisitions, HLC’s management will likely buy back stock

Cyclicality of the hotel business

The data below is from Smith Travel Research and the U.S. Department of Labor. Over time, ADR and RevPAR have done well compared to CPI, but one can note the cycles of rising ADR/RevPAR leading to a rise in supply, resulting in a down cycle, and so forth.

If this chart holds, we should be bracing for an increase in supply and another downturn. Will it happen? Is this U.S. data representative of what’s happening in Canada? Does this aggregate data apply to the primarily limited service hotels HLC owns? We must certainly make a note of this data; it would be great to be investing at the bottom of the cycle instead.

Risks, uncertainties, and why this may not work out

• This stock may never re-rate to a more appropriate cap rate, or it may be that it is already trading at its appropriate cap rate

• Our estimate of free cash flow could prove to be too high if there is a protracted downturn in the Canadian economy, or if the western hotels deteriorate further, or if we experience a downcycle in RevPAR, or all of the above

• Clarke may decide to sell its shares to another operator, depriving minority shareholders of a control premium

• Clarke may engineer a take under if the shares trade at a low price for a long period of time

• Excess supply may enter the market, depressing RevPAR (notably from Airbnb)

• The hotel business is cyclical and we may be investing in the wrong part of the cycle (we’re certainly not at the bottom, see above)

Catalysts

• More value-creating activities by management such as debt refinancing, hotel renovations, monetization of non-core assets, stock buybacks, acquisitions, developments on excess land, and a potential sale of the company.

Catalyst

More value-creating activities by management such as debt refinancing, hotel renovations, monetization of non-core assets, stock buybacks, acquisitions, developments on excess land, and a potential sale of the company.