Description

WSP Global (WSP) is an overvalued Engineering & Construction (E&C) firm based in Canada, whose feverish growth-by-acquisition spree and opaque accounting have obscured a deteriorating organic growth picture. WSP trades at about a one-turn premium to peer multiples, despite slowing organic growth, poor earnings quality (masked by aggressive earnings management), opaque accounting, a backlog of under one-year, recently overpaying for acquisition targets, a brand in the marketplace that is not premium, and pockets of outsized exposure to challenged end-markets (Canadian oil and gas, Australian mining).

WSP could conceivably trade as low as $14.00-15.00, or ~65% downside, over the next 12-18 months. Over the same time frame, we believe a reasonable risk case for our position is in the $42-48 range, which equates to a 4-to-1 risk/reward based on a current price of ~$40.

All numbers are in Canadian dollars, unless otherwise specified

Business Description and background

Over the last 5 years, WSP has grown its employee base at a staggering compounded annual growth rate of 50% (more than 3.5x its next closest peer), and its total asset base has ballooned by 9x. WSP has financed this torrid pace of acquisitions using debt and their own inflated stock, causing their share count to more than quadruple since 2010.

Canada’s two largest pension funds (Caisse de Depot and Canada Pension Plan) now collectively own 36% of WSP – as many of these equity offerings were private placements to these two institutional investors. These equity offerings have also provided WSP a powerful motive to manage earnings and promote their stock. WSP’s feverish M&A activity and equity offerings have also arguably compromised the independence of sell-side analysts covering the name.

One way we believe WSP has been managing earnings is by obscuring a slowdown in their topline through aggressive revenue recognition under percentage-of-completion (POC) accounting. This is evident in their bulging unbilled receivables-to-sales. WSP has masked the rise in their receivables, though, through the use of its own unconventional Day Sales Outstanding “DSO” metric. This metric (which IR could not tell me how the company calculates, and for which the provided disclosures are no help either) hides the extent of WSP’s rising DSO. In speaking with numerous sell-side analysts, we could not find one that is able to back into management’s metric. Also, according to channel checks we conducted and based on studying industry peers, we believe that there is a chance some of WSP’s receivables are bad, which if true should cause investors to question previously reported earnings levels.

Finally, WSP’s stock has benefitted from a broad reallocation of investment funds from resource to non-resource Canadian stocks, amidst the plummet in oil prices, which left much of the Canadian energy complex not investable for a broad swath of Canadian investors. This bubbly, market-wide reallocation of capital, coupled with excitement over a potential boost in infrastructure stimulus projects under the newly-elected Trudeau government, resulted in run-up in WSP’s stock price. All this excitement, however, has ignored or overlooked the poor quality of WSP’s earnings, and the significant execution risk inherent in their expansion strategy.

We believe investors are in for a rude awakening, as some of WSP’s earnings management catches up with them, the company’s organic growth slows in 2016, and they are unable to continue their inorganic growth due to a tighter capital markets environment.

Aggressive revenue recognition suggests possible cost overruns/revenue pull forward

At ~140 days, WSP’s DSO in the most recent quarter is dangerously high for an E&C company. When we calculate DSO using a conventional method, and use the same method for all E&C peers across Europe, US, and Canada, WSP has the highest DSO of any of its peers. About 75 days is a more typical, healthy DSO in the industry. An industry expert we spoke to said that a DSO approaching 150 for an E&C company is “go out of business” bad.

WSP, however, has obscured the severity of their problems by providing their own DSO metric – which registered an astounding “87” in the MRQ (and which is conveniently in-line with where industry peers should be). Also, management’s DSO metric is something that drives their compensation:

Further, management’s DSO compared to a conventional DSO calculation have been diverging in recent years. Management is possibly manipulating this metric to mask deteriorating operating performance and to maximize their compensation payout. It has been part of their Incentive Plans going back to 2011.

WSP introduced this disclosure in August 2013:

DSO is not an IFRS measure. It represents the average number of days to convert our trade receivables and costs and anticipated profits in excess of billings into cash. The Company's method of calculating DSO may differ from the methods used by other issuers and, accordingly, may not be comparable to similar measures used by other issuers.

In August 2014, the definition changed again slightly to:

DSO is not an IFRS measure. It represents the average number of days to convert our trade receivables and costs and anticipated profits in excess of billings into cash, net of sales taxes. The Corporation's method of calculating DSO may differ from the methods used by other issuers and, accordingly, may not be comparable to similar measures used by other issuers.

The use of an average number of days, while used by some other companies, is certainly not appropriate with a company that has rapidly growing sales, inorganic or otherwise. Also, the inclusion of “net of sales taxes” is not something we have ever seen in a DSO definition before.

In the MRQ, WSP’s cost and anticipated profits in excess of billings were $863mm (CAD), and net trade A/R were $1,433. Sales were $1,503mm in the Q. This amounts to a DSO of 138 vs. the 87 DSO that management reported!

There has been a balanced rise in billed and unbilled receivables (i.e. cost and anticipated profits in excess of billings) over the last year. In December 2014 (the first full Q to include Parsons), unbilled A/R was only $594mm and trade receivables were $1,026mm. Going back to September 2012, though, the rise in unbilled receivables has been much more pronounced than has the rise in trade A/R DSO, suggesting either cost overruns or aggressive revenue recognition under percentage-of-completion accounting.

Given WSP maintains a backlog of less than one year, and their organic growth is decelerating (was negative in MRQ, “Globally, we experienced negative organic net revenues of 1.8%, mainly as a result of the downturn [in Western Canada oil and gas sector]. Excluding the drop in this sector, organic growth amounted to 4.9%, in line with our 5% objective."), one would anticipate they should be freeing up working capital.

Instead, working capital has been a drain on cash as of late:

Over the LTM period, reported LTM CFFO was only 56% of Net Income plus Depreciation and Amortization and Stock-based Comp. A huge earnings quality red flag.

Reported LTM FCF was $80mm ($4 bln market cap), or 2% FCFE yield.  

Note also that WSP’s cash flow statement does not yet include acquired accounts from Parson’s in their cash flow reconciliation, and will include this for the first time next quarter. We believe WSP’s FCF will worsen once it includes reconciliations of Parsons’ accounts:

This combination of variables suggests declining, or at very least struggling, contract win rates, integration struggles with past acquisitions, aggressive revenue recognition to maintain illusion of growth, softness in some of WSP’s key end-markets, and potentially significant cost overruns in key fixed-price contracts (which represent ~50% of all contracts), and/or bad receivables that WSP has delayed writing-off.

Backlog < 12 months of Revenue

Despite the fact that WSP has been growing rapidly through M&A since 2011, they have persistently operated in recent years with a backlog in the range of 8-10 months of revenue. For an E&C company, whose primary expense is labor, this is not a sustainable path. To give you a sense, ACM’s backlog is currently 2.2x LTM revenue. FLR is 2.3x. JEC 1.5x. In a broad peer universe, there are only two companies with a lower backlog relative to LTM revenue than WSP (Fugro NV – which has huge oil and gas exposure and Stantec, the Canadian company that WSP often gets benchmarked against).

WSP’s stock price performance vs. Fugro NV and Stantec since beginning of 2015

WSP does also disclose a “soft backlog”, but given this is work “for which funding may not have yet been designated”, we think it should be ignored.  

It is generally bad business to pay engineers and consultants 12-month, full-time salaries, while only mustering a pipeline of 10-months of work (especially when your DSO is over 4-months). With this setup, and in the absence of rapid growth in their backlog, WSP is destined for suboptimal labor utilization, and declining margins. To-date WSP has furiously “scrambled the egg” through acquisitions, such that these simple economics have not yet caught up with their stock price. But it is tough to envision how it will not catch up to them eventually, like it has for their peers with similarly small backlogs.

Recent Acquisitions done at hefty multiples and huge Execution risk

On 10/31/14, WSP completed its acquisition of Parsons Brinckerhoff from Balfour Beatty. BBY LN sold Parsons as part of a larger restructuring effort to shed assets, improve liquidity, and to turn around their operations, which have been hobbled by cost overruns among other issues.

About five years prior to selling Parsons to WSP, Balfour paid $626mm USD for Parsons. WSP paid $1,242 USD for Parsons, or 8.8x EV/EBITDA. There is little chance that BBY LN doubled the value of the Parsons business, while it was under their ownership. We spoke to an experienced industry executive who felt that acquisitions at any multiple above 6.5x EV/EBITDA in this environment is overpaying. Also, WSP paid this multiple to a distressed and largely forced-seller. WSP also paid 8.9x for MMM, which closed in October 2015. WSP’s peer universe trades currently for ~7.5x EV/EBITDA. The synergies that exist in these deals are largely just the consolidation of redundant office spaces, and some corporate overhead.

Further, regarding Parsons, WSP decided to keep the WSP brand in their name, rather than to fully assume the Parsons brand. One industry executive we spoke to suggested the Parsons brand had far more cache in the marketplace than WSP. Because of WSP’s origins as a rather small, obscure Canadian company called Genivar, and the way in which they rocketed on to the scene by aggressively overpaying for and rolling-up peers, the WSP brand (which Genivar assumed when they took over WSP in 2012) is not as well-known in the marketplace. Also, the initial rebranding to WSP was for inauspicious reasons to say the least - Genivar’s involvement in a political corruption scandal.

The implications of WSP not fully assuming the Parsons brand is that the Parsons’ book of business is not necessarily going to translate fully over to WSP under the WSP brand, which is a scary possibility given WSP paid such a hefty multiple for Parsons.

Also, the sheer size and pace of WSP’s M&A activity is fraught with execution risk, as WSP by our estimate has added some $3.6 bln in revenues via acquisition since June 2012 (vs. FY16E sales of $5 bln).

FX tailwinds abating

WSP has recently benefitted from translating foreign earnings back to CAD, given significant weakening in the CAD. But this benefit peaked in MRQ by our estimate, and will start to diminish y/y in 2 Q’s. This benefit along with WSP’s inorganic growth has been helping to mask WSP’s weakening organic growth trends. Here are our estimates of the FX tailwinds:

Now, it is very hard to get too precise with these estimates, because the company has not yet provided very detailed information about sales by geography for the pro-forma enterprise after their recent acquisitions – and recent acquisitions have made WSP a more global company.

But one thing we can do is back into what the street is implying as the FX tailwind for FY15, based on our own estimate of revenues added from acquisitions and by plugging the organic growth targeted by company, for which the Street is largely given WSP credit for achieving.

We estimate in 2015, acquisitions will add $1,726 CAD in revs. Street is calling for $4,414 in FY15 vs. $2,350 year prior. Company is targeting 14% from FX and organic ($2,688 over $2,350), if we assume they achieve their 5% organic in 2015, this would imply 9% tailwind to legacy business, which is a little high to what we estimate as the tailwind for FY15. Also, recent commentary from management suggests 5% organic growth in Q415 is not likely. So either Street is likely overstating the FX benefit WSP will realize in Q415, or WSP’s organic growth, or some combo of both.  

Looking at 2016 – Street is calling for $4,979mm in revs – or 13% growth y/y from FY15. We estimate $246mm of will be inorganic growth from recent M&A.  So that is ~7% organic + FX. I estimate at current FX rates that the FX tailwind would be ~4% for FY16. So that would imply organic revs of ~3% (which is below company target of 5% - but by our analysis way too high). Once Parsons is included in their organic growth – and as their DSO shenanigans catch up to them – WSP’s organic growth will likely turn negative (and also organic growth + FX tailwind in 2016 will be well below the consensus estimate of 7%).

Non-resource Canada bubbling over

36-months ago, with WTI Crude just below $100, investment dollars were pouring into resource Canada, meanwhile Industrial Services companies in Canada traded at a comparable median valuation multiple as their U.S. counterparts – see table below. But soon thereafter Crude prices peaked, and as the price for oil began to plummet, the stock prices of many Canadian resource companies naturally followed suit. At the same time, as Canadian banks and investors repurposed capital out of resource and into non-resource Canada, valuations for the latter group inflated, particularly relative to their European and U.S. counterparts.

For example 36 months ago, the median LTM P/E of Canadian and U.S. Industrial Service companies were 16.7 vs. 15.9 respectively, fairly in-line. 24 months ago, as price for Crude started to fall, Canadian LTM P/E had ballooned to 24 vs. 18.9 in U.S. Currently, Canadian LTM P/E is 17.6 vs. 14.6 for U.S., and 10.5 for Europe – so still very inflated relative to these two regions

Some have cited the promises of the Trudeau government for more infrastructure spending as a reason to be more bullish on Canadian E&C companies. But the economic and political realities facing the new leader of Canada are already forcing a dampening of earlier infrastructure spending plans. Also, WSP has become an increasingly-global company with exposure to more troubled pockets, like Australian mining companies for example. The other reason for the increase in Canadian valuations relative to U.S. and European peers is perhaps due to investors extrapolating unsustainable FX translational tailwinds.

Regardless of these bullish arguments, we believe that the valuation premium of Canadian Industrial Services companies to their U.S. and European peers is unjustified, and simply the function of excess capital pouring out of resource Canada and looking for a place to land, rather than it is due to any valid, sustainable macroeconomic forces.

Non-GAAP Metrics opaque – and mask potential earnings shenanigans

Many companies are making increasingly aggressive use of non-GAAP and non-IFRS adjustments, a trend the WSJ highlighted in a recent article called The Widening GAAP Gap.

So that WSP is one more company of the many that is guilty of a more liberal use of adjustments is perhaps not big news. What does make WSP unique, though, is the high degree of opacity of their adjustments.

Specifically, WSP lumps all their non-GAAP adjustments into a catch-all of “non-underlying items” with no great disclosure as to what those “non-underlying items” are exactly. This ambiguity gives WSP incredible wiggle room for financial shenanigans.

This non-IFRS, catch-all measure first started appearing in September 2014, and represents all the company’s earnings growth for FY14:

Here is their disclosure on “non-underlying items” from the Glossary:

Non-underlying items is not an IFRS measure. Non-underlying items are items of financial performance which the Corporation believes should be excluded in understanding the underlying financial performance achieved by the Corporation. Non-underlying items include transaction and integration costs related to business acquisitions as well as costs of restructuring and reorganizing existing operations. In Q2 2015, non-underlying items included a gain made on the disposal of an equity investment in an associate.

Insider sales and insider behavior

Even as WSP has been aggressively issuing equity to fund overpriced, mammoth-sized acquisitions – the CEO, Pierre Shoiry, has also recently made some chunky, very well-timed open market sales. The first sale was for 100k shares in September 2015, the second for 169.8k shares in November 2015. The November sale almost perfectly top-ticked the stock price. Together, these sales comprised ~30% of his WSP holdings, as Shoiry now owns 616K shares.

Further, it is worth mentioning that WSP and its leadership have been embroiled in untoward behavior in the past. In 2012, when the company was still largely just a Canadian company called Genivar, they were a subject of a public inquiry (called the Charbonneau Commission). This commission was enacted in late 2011 and was tasked with looking into the potential corruption in the management and awarding of public construction contracts. “Witnesses…detailed a system of bid-rigging that saw a cartel of engineering and construction firms obtain public contracts from the city of Montreal in exchange for political donations…engineering companies that were part of Longueuil’s system at the time were Genivar Inc., Dessau, Groupe SM and Cima+.”

This likely factored in to the company’s desire to acquire their way out of the Genivar identity and to change the corporate name. The current CEO and CFO of the company were also in charge during the time of the alleged bid-rigging.

Valuation

Current multiples; Historical; Peer

• EV/NTM EBITDA: 7.7x; 8.2x; 7.5x

• P/NTM FCF: 13.2x; 13.8x; 13.4x

• P/NTM EPS: 14.1; 15.3x; 13.0x

Scenario Analysis Assumptions – Apply NTM multiples to 2018 results as of 12/31/17, and take an average of the three valuation methodologies.

Base case (’18 Target Price $CAD $14.65; IRR 40%):

• EV/NTM EBITDA: 7x   

• P/NTM FCF: 13x                            

• P/NTM EPS: 13x                                  

Assumptions: Estimate as much as 17% of 2015 revenue could be related to aggressive revenue recognition. Organic net revenue slows to -2% as pull-forward of revenue comes back to haunt them, and end markets soften. Other operational costs grow 2%, and personnel costs per employee-to-revenue per employee creeps up to 81% from ~76% (i.e. decline in labor utilization). EBITDA margins 2018 decline to ~6.1% from ~10% in ‘15.  

Downside case (’18 Target Price $CAD $46.79; IRR 8%):

• EV/NTM EBITDA: 8x    

• P/NTM FCF: 15x                           

• P/NTM EPS: 14x                                  

Assumptions: 24% tax rate; 7% organic growth; no new acquisitions; no new debt or equity offerings; no deterioration in margins (10.4% 2018 EBITDA%). Operational costs grow 300 bps > organic growth. Personnel costs per employee-to-revenue per employee of 77%. No write-offs or one time charges. All receivables are healthy.

Risks

• Macro headwinds subside or reverse: Mining, Oil & Gas, China Buildings, Middle East projects, Australia, Canada.

• U.S. Infrastructure spending dramatically increases; and a rising tide lifts all boats.

• Management is able to continue to use cheap currency (i.e. inflated stock price) to make further acquisitions that “scramble the egg” further.

• AECOM (like they did with URS) buys WSP.

This report (this “Report”) on WSP Global (the “Company”) has been prepared for informational purposes only. As of the date of this Report, we (collectively, the “Authors”) hold short positions tied to the securities of the Company described herein and stand to benefit from a decline in the price of the common stock of the Company. Following publication of this Report, and without further notice, the Authors may increase or reduce their short exposure to the Company’s securities or establish long positions based on changes in market price, market conditions, or the Authors’ opinions with respect to Company prospects. This Report is not designed to be applicable to the specific circumstances of any particular reader. All readers are responsible for conducting their own due diligence and making their own investment decisions with respect to the Company’s securities. Information contained herein was obtained from public sources believed to be accurate and reliable but is presented “as is,” without any warranty as to accuracy or completeness. The opinions expressed herein may change and the Authors undertake no obligation to update this Report. This Report contains certain forward-looking statements and projections which are inherently speculative and uncertain.

Catalyst

• Write-off of bad receivables. Bear in mind that this should be thought of as admitting to overstated financial results in the past, rather than a one-time hit to the balance sheet.

• Parsons results begin to be included in organic numbers (Q415)

• Organic growth declines further

• Personnel costs per employee-to-revenue per employee spikes amid sluggish backlog

• Further Insider sales

• Further backlog declines