PPHE Hotel Group — Going up

The half to June was notable for completed repositioning of Holmes Hotel London and Arena Kažela Campsite in Croatia and soft openings of repositioning projects, Vondelpark Amsterdam and Utrecht (both to be fully operational in H219).

As shown in Exhibit 1, London stole the show with EBITDA estimated to be up c 8% despite Holmes disruption. We estimate the core estate to have increased RevPAR by c 8% against market 5%. Although this is creditable, it is flattered somewhat by inclusion of newly repositioned Riverbank and apparent market consolidation in H118 – see Exhibit 2. We assume a strong contribution also from maturing assets, Waterloo and Park Royal.

By contrast, Netherlands disappointed as the impact of an unexpected number of rooms off at Vondelpark and Utrecht obscured the positive returns from newly completed investment at Victoria Amsterdam. Consequently, EBITDA was flat on 4% higher revenue, ie margin down from 29% to 27%.

A small improvement in Germany and Hungary EBITDA resulting from termination of an unprofitable lease at art’otel dresden was countered by a return to off-season loss in Croatia attributable to poor weather in May, cost inflation and a tightening market.

The outlook for the rest of this year is broadly positive with PPHE reporting both London and the Netherlands to be continuing to trade well. Most significantly, the London market may be expected to return to more normal levels of RevPAR growth, we expect 2–3%, after almost 5% in H1. Security and Brexit uncertainty are concerns, but London has shown admirable resilience and benefits from increased tourism owing to sterling weakness. Greater measurable worries are room supply, which is above its long-term trend, slower economic growth and rising operating costs. GL, London's largest hotel owner/operator, newly confirms that it 'maintains a cautious outlook', whereas, per pwc, 2019 market RevPAR gain may be marginal. Our H2 RevPAR forecast for the core London estate is +2% in view of a satisfactory 3% market gain in July and August and likely initial returns from the £9m Holmes investment, offset by a challenging Q4 comparative (+10%; see Exhibit 2). There should also be continued double-digit percentage growth in EBITDA at maturing Waterloo and Park Royal. We are therefore raising our full-year UK pre-IFRS 16 EBITDA forecast from £65m to £69.4m.

In contrast, we are downgrading underlying Netherlands full-year EBITDA from £16.5m to £14.8m, given the aforementioned project delays. This is still ahead of 2018 and, encouragingly, it assumes a 10% improvement in second-half EBITDA on full reopening of Vondelpark and Utrecht in contrast to the flat outturn of H1.

Croatia is also set to disappoint, with a likely absolute reduction in profit. We now look for 2019 EBITDA of £16.6m against £18.6m in 2018 and our original forecast of £20m. The first-half dip aside, the shortfall largely reflects increased competition owing to sustained recovery in neighbouring tourist markets, notably Turkey, eg Antalya foreign visitors up 18% YTD according to the Turkish Ministry of Culture and Tourism, as well as inflationary pressures. An apparently encouraging 5% increase in foreign tourist nights (Exhibit 3) in Pula, the centre of Arena / PPHE operations, in June and July, per the Croatian Bureau of Statistics, is arguably misleading as June 2018 was subdued by the football World Cup and July, much the more important month, was marginally lower, which bodes ill for the other key trading month, August. On a more positive note, we expect gains from the maturing of the £8m Arena One 99 luxury glamping development (awarded best tourist product of 2018 by Istria Tourism Board) and the repositioning of Kažela campsite with new infrastructure, facilities and premium mobile homes.

Solid core trading and investment payoff, newly endorsed by the prestigious award of AA Large Hotel Group of the Year 2019, should again favour next year, hence our continued expectation of a further mid-single-digit rise in EBITDA. Costs may be expected still to be an issue (not only in London, as well known, but in Croatia). As for 2019, the £5m upgrade in our headline EBITDA is due wholly to IFRS 16 adoption. A profit catalyst would be expansion although returns from a development project would necessarily take time to accrue.

The June balance sheet confirmed finances are in good shape. The sharp rise in net bank debt in H119 (£524m up from £480m) reflects pre-construction costs for art’otel new york (£14m) and sustained high capex (£43m) in particular. However, this is well within long-term bank facilities of c £700m, which have an average term to maturity of 7.5 years, interest cover of 4x and average cost of 3%. Acquisition apart, we envisage only a small increase in year-on-year net debt at December 2019 (£506m). With regard to expansion, PPHE has a well-established policy and track record of borrowing within a 65% loan-to-value requirement of its banks. Management is confident it can leverage on its assets by opening similar facilities for future projects.

PPHE’s current share price is visibly below its last reported EPRA NAV per share of £25.52. We have thus analysed whether this discount to NAV is justified by the fact that the company is not a pure-play property investor as it manages hotels and operates some under operating lease agreements. From a valuation perspective, we believe it is instructive to examine the average capitalisation yield on PPHE’s property portfolio implied by the current share price. This is an important parameter for investors as it measures the attractiveness of PPHE’s current market valuation from the perspective of hotel earnings afforded by the company. Our first step is to estimate hotel net operating income (NOI), based on reported group EBITDAR (earnings before interest, taxes, depreciation, amortisation and rental expenses) on a last 12-month basis at end-June 2019 excluding earnings from the management and central services division (see Exhibit 4). For a detailed discussion of our methodology, please refer to our previous outlook note.

Our next step is to divide NOI by the implied value of PPHE’s hotel properties based on the current share price. To arrive at the implied value, we take the market cap and add the last-reported net debt figure (at June 2019) after excluding the proportion attributable to Arena minority shareholders. We have also excluded PPHE’s non-real estate assets and liabilities (net working capital items in particular) and the value of the Hoxton project (£82.5m according to the recent valuation by Savills).

To conclude, we have estimated the value of PPHE’s hotel management business to isolate the value of its hotel properties. These activities are represented by the management and central services division, which generated £43.8m revenue (before eliminations) and £11.2m EBITDA for the 12 months ended June 2019. We have valued the business using a DCF method based on a WACC (calculated separately for each year) of 7.3–7.8% and a terminal growth rate of 1.5%, arriving at a fair value of £165m. We have allowed for art’otel hoxton but not for the two new art’otel management contracts. This clearly illustrates the management business represents incremental value on top of the NAV of PPHE’s properties.

As a sense check to our DCF valuation of management and central services, we have conducted a comparative valuation based on global peers with relatively asset-light models, including Wyndham Hotels & Resorts, IHG, AccorHotels, Marriott and Hilton. Our estimated average EV/EBITDAR ratio adjusted for capitalised operating lease liabilities for this peer group is 14.8x for LTM ended June 2019 (used in the absence of available EBITDAR consensus). When applying this multiple to PPHE’s management and central services division’s LTM EBITDA (adjusted for refurbishments and asset ramp up), we arrive at a fair value of £169m. We acknowledge that our selected peers possess a much stronger brand and market presence than PPHE. However, as our DCF-derived fair value also captures growth prospects associated with the new management contracts, we include the average for both methods in our yield calculations at £167m (390p per PPHE share, see Exhibit 5).

Therefore, we obtain an implied capitalisation yield of 6.6%. We estimate the prevailing weighted average prime capitalisation yield in the property markets where the company operates is c 5% and thus visibly below PPHE’s implied capitalisation yield, suggesting the market is still not fully reflecting the value of the company’s properties. We also estimate that the current average implied capitalisation yields for US hotel REITs (used in the absence of listed European hotel REITs) stands at 8.4%, which is higher than our estimate for PPHE. However, we note that market yields for major central business district locations in the US are higher than market yields in London.

An alternative way to look at PPHE’s valuation is to examine its return on EPRA NAV of £25.52 per share as implied by the company’s EPRA EPS, which was just 4.7% at end 2018. However, we note that PPHE’s EPRA NAV includes the Hoxton project, which has yet to generate profit. Moreover, PPHE has a high cash balance while looking for projects for acquisition. The company estimates that after adjusting for these, the return on EPRA NAV is 5.5%. We also note that PPHE’s financials remain affected by renovations. Based on our EPRA EPS forecasts and assuming a flat EPRA NAV, the return increases to 5.9% in 2020 and 6.2% in 2021. Indeed, since the current share price is below EPRA NAV per share, we have also calculated the implied return on the market cap (based on the company’s adjustments for excess cash and Hoxton project), which stands at 9.0% based on LTM earnings (as at end-June 2019). This confirms our conclusions from implied capitalisation yield analysis.

Based on the above calculations as well as Savills/Zagreb Nekretnine’s EPRA NAV estimate, we may conduct a sum-of-the-parts valuation (see Exhibit 7). After adjusting for net debt, minorities and non-real estate assets, we obtain a value of 2,964p per share. Importantly, these calculations are based on the assumption that the market is applying no discount to the NAV valuation of the properties. However, we acknowledge that US hotel REITs have been recently trading at a discount to NAV of around 5–20%. Therefore, if we were conservatively to apply a 20% discount to PPHE’s net property value, we would arrive at 2,453p, which still provides material upside to the current share price.

In summary, our implied capitalisation yield analysis and the SOTP valuation indicate that PPHE’s discount to EPRA NAV is not justified by its management business (which we believe to be a clear value contributor), while activities on operating leases are negligible in terms of valuation. This is confirmed by our propco-opco model (see our previous Outlook note for details), which still indicates to a small positive value of the hypothetical stand-alone entity operating PPHE’s hotels under a lease agreement (opco) at 27p per share (based on a conservative rental yield of 5.5%). As most properties are in prime locations and well invested, we believe PPHE constitutes a good entry point for exposure to the London and Netherlands property market at an attractive yield level. See our previous outlook note for a detailed asset quality review.