PPHE Hotel Group — Going places

In the face of exceptional security challenges in its London heartland, PPHE showed remarkable resilience and focus to combine decent financials with active, uninterrupted implementation of a busy development schedule. This included a material addition (over 700 rooms) to the London portfolio (now c 3,150) with the openings of Park Plaza Waterloo and Park Royal (fully operational by period end) and participation in a transformative €106m fund-raising by its Croatian subsidiary, Arena Hospitality. There was also minor adjustment to the estate in Germany (acquisition of two Berlin properties previously under operating leases) and the Netherlands (sale of part of Park Plaza Vondelpark Amsterdam). Last but not least, at the very end of the half, the company entered into a sale and leaseback agreement for Waterloo.

Despite striking increases in H117 headline revenue and EBITDA of 27% and 23%, as shown in Exhibit 1, a lot of moving parts makes it incumbent to identify underlying performance. Therefore we begin by adjusting to omit recent openings in London and Germany, to include Croatia’s pre-acquisition Q116 and to reflect the new freehold position of two Berlin hotels. These temper respective like-for-like advances to 16% and 21%. We make further revision for currency in view of sterling depreciation. Adjusting for this translation boost reduces revenue and EBITDA gains to 11% and 18%. This double-digit growth is still impressive and even more so as broadly spread, ie RevPAR up c 10% and 18% in London and Germany and Hungary, respectively, revenue +19% in Croatia and RevPAR up 4% in the Netherlands despite a strong comparative and renovation.

It barely needs pointing out that London’s aforementioned apparent bumper performance should be seen against a half subdued by the November 2015 attacks in Paris when a 5% reduction in RevPAR was in line with market reports for the capital – see Exhibit 3. Moreover, the company’s 10% yield gain in the period was no more than that of the market. However, it is much to the credit of London and PPHE that such positive recovery was achieved in the face of further major attacks (Westminster Bridge and London Bridge), especially given their proximity to the company’s South Bank focus. Pound devaluation following the Brexit vote has provided a signal boost to tourism and thus the hotel sector. According to VisitBritain, visits to the UK were up 9% in H117. While arrivals from the EU (two-thirds of visits) saw a rise of 4%, there was notable demand from North America (up by a quarter) and the rest of the world (+18%).

While London may thus have been flattered in H1, the reverse applied to Croatia as its off-season Q1 was included for the first time and risks obscuring exceptional buoyancy (revenue +35%) in June, the first high-season month of the year. If Q116 is included for comparability, the increase in like-for-like local currency revenue was almost 20%. Q217 alone was up 25%, albeit boosted by a late Easter. As shown below, such progress reflects a step-change in the appeal of Croatia as a tourist destination in addition to payoff from the company’s significant pre-season investment.

The performance in the Netherlands is similarly deserving of explanation as a reported decline in revenue, however minor (2%), is at odds with a business that has seen consistent growth. Fortunately this is simply attributable to renovation at the flagship Victoria Amsterdam and reduction in size of Vondelpark Amsterdam (we have assumed 50 rooms off or c 4% of total inventory). 4% RevPAR gain was respectable against a buoyant H116 (5% ahead).

Germany and Hungary were also somewhat out of kilter, with RevPAR up an exceptional 18% l-f-l in the period. This reflected recovery from a soft first half in 2016, depressed by renovations and market weakness.

Like-for-like notwithstanding, first signs from Waterloo and Park Royal openings were encouraging and Nuremberg, which opened in H116, looks to be maturing well. Waterloo in particular has filled up quickly, capitalising on the company’s unique South Bank concentration and general market recovery. Park Royal, whose launch was delayed into Q2, has more the trading characteristics of a UK provincial hotel.

Although like-for-like EBITDA improvement of 18%, as discussed above, is per se no mean feat, such all-round buoyancy might have been expected to deliver yet higher returns by a business with such operational gearing. Total RevPAR gain was, after all, 18% l-f-l and room rate-led, albeit currency boosted.

However, the UK, much the company’s principal profit centre, saw accelerating pressure on labour supply and operating costs as well as increased security expenses following recent attacks. In addition to a persistent hike in business rates, sterling weakness accounted for a rise in imported food costs and a shortage of qualified staff as remittances by migrant workers were no longer worth so much when sent to their families back home. Specifically, to show the scale of the challenge, management has commented on a shortage of 90 housekeeping personnel out of a team of 400, resulting in their payroll costs now somewhat ahead of national minimum wage rate. For the core London hotels, we have therefore assumed for H117 only maintained EBITDA margin despite double-digit top-line growth (EBITDA is disclosed only at country level).

By contrast, Croatia did indeed convert successfully. On a like-for-like basis, ie including pre-acquisition Q116, a 19% rise (HRK20m) in local currency revenue led to the elimination of HRK8m EBITDA loss.

The shortfall in Netherlands EBITDA was due simply to rooms off and associated disruption from renovation, as mentioned above. 2% lower local currency revenue resulted in an 8% decline in EBITDA, ie margin down from 32% to 30%.

The outlook for the rest of this year and 2018 remains one of sustained underlying growth in all PPHE’s operating regions. Most significantly, London may be expected to return to more normal levels, say +2-3% RevPAR. Geopolitical events, eg security and Brexit uncertainty, are necessarily a concern but the hotel market has shown admirable resilience and is a beneficiary of increased tourism as a result of sterling weakness. A greater measurable worry is room supply, which is set to rise above its long-term trend over the next year. Already market occupancy has fallen in each of the last three months, according to STR, even though demand continues to go up. Such was the impact in August of a spike in supply (4% against 2.7% in 2016 and 2.2% the year before) that the month’s RevPAR was merely flat. In the light of this more modest revenue growth and persisting cost issues (labour and imported inflation) we expect further EBITDA margin erosion in London (excluding openings) in both H2 and next year.

Similar RevPAR growth to the UK levels is on the cards in the Netherlands but the renovation impact will intensify with up to 20% of inventory off; we have assumed 200 rooms off in the current half and 100 in 2018, predominantly at the 299-room Victoria Amsterdam. As a result, we forecast local currency H217 EBITDA down by almost 30% but recovering slightly next year on initial investment payoff.

Croatia is arguably the joker in the pack as our +3% RevPAR forecast for both periods seems cautious, given the company’s flourishing June, enthusiasm about summer trading and the Croatian Bureau of Statistics’ confirmation of market strength in July (foreign visitor nights up 11%, giving cumulative June-July +18%, as shown in Exhibit 3). We are also encouraged by likely returns from increased investment and extension of business, eg music festivals and sporting competitions, in the resort hotels’ shoulder months.

We remain positive about the London openings, which achieved a better than expected 18% EBITDA margin in their initial half.

Given broadly unchanged assumptions, adjustments to forecasts for both periods are minor but positive at the revenue and EBITDA level, ie respectively 3% and £4m in 2017 and 5% and £3m in 2018.

The reduction in our PBT forecasts (£2.5m this year and £4m next) is due to higher depreciation than we had expected (likely up 35% this year, driven by the London openings) and finance costs inflated by Waterloo rent after the sale and leaseback. In any case EV/EBITDA remains our preferred valuation metric.

The June balance sheet confirmed finances to be in good shape. While net debt remains well within long-term bank facilities (average term to maturity of eight years), of particular note was record cash of c £190m, of which c £100m was in Arena after its public offering. To this should be added c £80m from the subsequent Waterloo sale and leaseback.

Last year the company recognised that after its extensive refinancings it had excess liquidity, so paid a special dividend of 100p, ie c £42m. By contrast, a repeat has now been formally deferred for 12 months.

It was thus no surprise that management has confirmed that it is looking actively to make investments, while also prepared to bide its time for opportunities when the market subsides. The current focus is both single London properties and UK portfolios. Office conversions on the Waterloo model (previously Hercules House Government offices) are also being examined.

Given management’s successful development record, this is particularly exciting. Indeed, the investment case of Waterloo alone is persuasive. An independent valuation of £250m revealed a doubling of value in just four years. Otherwise we could point to similar appreciation of PPHE’s 52% investment in Arena, valued at €161m at end August against a total cost of €79m.

We are also enthusiastic about potential returns from investment by Arena in both Croatia and central Europe.