PizzaExpress:Q3earnings,2out of 3ain't bad


PizzaExpress’s Q3 UK LFLs were negative as we had feared, but adj. EBITDAwas a beat to our estimate and margin performance was consistent withmanagement guidance. Despite lowering our FY18/19 estimates below and apositive reaction in the bonds today, we continue to like the risk-reward andreiterate CreditBuy on both the Senior Secured (96/7.9%) and Senior(89/11.7%) notes. For the quarter ended 1st October, sales of £140m were up3.6% yoy, but UK & Ireland lfls slipped to -1.3% (at cFX) from +0.9% in H1reflecting a challenging market in London in particular. International lfls werealso soft at +0.7% (V H1 5.0%), but according to management this largelyreflected the later timing of China’s Mid-Autumn Festival in Q4 this year.

Maintain Sell on Lenovo 2022s.

    Group Adj. EBITDA of £25.5m was down £1m yoy but ahead of our £24.7mest., as a £2m decrease in the UK & Ireland (£23.3m) was partly offset by a£1m increase in the International business (£2.1m). As guided bymanagement, the UK & Ireland EBITDA margin deficit improved from -460bp(yoy) in H1 to -190bp in Q3, reflecting the termination of dual running costs forthe new EPOS system as well as efficiencies which offset organic inflation(e.g. labour, rates, ingredients) of c. 300bp. Compared with Q2, net debtincreased by £16m to £645m mainly reflecting coupon payments whileSenior/Sub net leveraged increased to 4.7x/6.8x from 4.5x/6.6x. Managementupdated site opening plans with the UK likely to fall to just six next year from c.

    Our regular readers will know that our negative view on Lenovo 2022s hasbeen driven by concerns around underlying businesses and rich valuations.

    12 in FY17, while International will increase to c. 40 from c. 32 this year; weest. total capex will be £44m this year and £42m in FY18. The company alsodisclosed that it sold the Firezza business, which comprised 22 mainly takeawaysites in London and was making a negative contribution to LTM EBITDA.

    The former hasn't abated yet. Although the HKD3.9 billion equity raising in Octcomes as a positive surprise, it is unlikely to see immediate deleveraging giventhe announced acquisition of 51% stake in Fujitsu’s PC business and itsplanned investment of USD1.2 billion in AI over 4-5 years. In addition, the putoption for Compal to offload its 49% interests in the JV (up to USD750 million)to Lenovo is exercisable after Oct 1, 2017. From a valuation perspective, weacknowledge bonds don't look as stretched now. The 2022s (G+185bp bid)have underperformed, widening ~5bp since issuance in mid-March, against~30bps tightening for China Corp during the same period. However, comparedto other unrated bonds from China like Chalco 2022s (G+175bp) and ChinaChengtong 2022s (G+170bp), the Lenovo bonds still don't look cheap. Evenlooking at higher BB rated HK names like Lifestyle 2022s (G+205bp), Lenovoappears expensive. The perps ($102.75 ask, G+275bp) on the other hand wasrelatively fair and we stay Hold on them. We do note though that these perpshave a weak structure with just a "performance guarantee" from Lenovo.

    We were generally impressed by new CEO Jinlong Wang who unsurprisinglyemphasised the growth opportunity in China where he was previously actingCEO, but also seems well-acquainted with the challenges in the UK. On theoutlook, management indicated negative lfls and margin erosion (c. 200bp) willpersist in the UK in Q4, but also indicated growth in China should broadlyoffset UK weakness next year. In our updated forecasts below, we see adj.

    Acquisition of 51% stake in Fujitsu’s PC business.

    EBITDA of £92m both this year and next, with operating cash flow moderatelynegative at -£5m to -£8m and Senior/Sub net leverage remaining high at4.6x/6.8x. However, we believe the core credit story remains intact with astrong core brand and growth in China likely to make a material contributionahead of the 2021 bond maturities.

    Lenovo unveiled its plans to acquire 51% stake in Fujitsu Client Computing Ltd(FCCL) while Fujitsu would retain the remaining 49%. Total consideration is~HKD1.23 billion, plus performance-based payment of HKD175 million to HKD875 million payable shortly after March 2020. Fundamentally, FCCL generatedpositive operating profit (HKD633 million for the year ended March 2017) andits asset size is relatively small at HKD10.3 billion at Mar’17. Operationally, thedeal will help Lenovo to increase its market share in Japan to ~40% from~30% per Bloomberg. While certainly not negative, we don’t foresee anyfundamental positive change in Lenovo’s underlying businesses (PC'sstructural challenges and Mobile's losses) due to this deal alone.

    2Q results.

    - EBITDA reached USD240 million (+10% yoy/+52% qoq) on revenues ofUSD11.8 billion (+5% yoy/+17% qoq). The margin improvement is mainlysupported by sacrificing growth for profitability in the PC segment (~70% ofrevenues) with 1.5% yoy decline of PC shipment, and ~25bp expansion in pretaxprofit margin. New smartphone launches brought a 19% qoq revenueincrease in the Mobile segment, but also leaded to higher advertising andpromotion expenses and thus softer pre-tax numbers in this segment (-USD132 million in Q2 vs -USD129 million in Q1). Data Center Business (~8% ofrevenues) continued to be the weakest segment by profitability, with pre-taxprofit margin at -10%.

    - Adjusted OCF was positive at USD346 million, versus negative USD568million in 1Q17 and positive USD1.6 billion in 2Q16. Capex stood at USD173million during the quarter, excluding stake sale of USD82 million and assetsdisposal gain of USD5 million. Overall FCF was slightly negative at USD32million after dividend payment of USD292 million.

    - On the balance sheet front, total debts and cash balance came down toUSD3.7 billion and USD1.3 billion, respectively, from USD3.9 billion andUSD1.5 billion at Jun’17. LTM gross leverage decreased slightly from 4.4x to4.1x while LTM net leverage was a tad higher from 2.6x to 2.7x.

    Please see Figures 1 to 7 for key financials.