DFAN14A

 

 

UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

WASHINGTON, D.C. 20549

 

 

SCHEDULE 14A

(Rule 14a-101)

SCHEDULE 14A INFORMATION

Proxy Statement Pursuant to Section 14(a) of

the Securities Exchange Act of 1934

 

 

Filed by the Registrant  ☐                             Filed by a Party other than the Registrant  ☒

Check the appropriate box:

 

  Preliminary Proxy Statement
  Confidential, for Use of the Commission Only (as permitted by Rule 14a-6(e)(2))
  Definitive Proxy Statement
  Definitive Additional Materials
  Soliciting material Pursuant to §240.14a-12

Automatic Data Processing, Inc.

(Name of Registrant as Specified In Its Charter)

William A. Ackman

Veronica M. Hagen

V. Paul Unruh

Pershing Square Capital Management, L.P.

PS Management GP, LLC

Pershing Square, L.P.

Pershing Square II, L.P.

Pershing Square International, Ltd.

Pershing Square Holdings, Ltd.

Pershing Square VI Master, L.P.

(Name of Person(s) Filing Proxy Statement, if other than the Registrant)


Payment of Filing Fee (Check the appropriate box):

  No fee required.
  Fee computed on table below per Exchange Act Rules 14a-6(i)(4) and 0-11.
  1)  

Title of each class of securities to which transaction applies:

 

     

  2)  

Aggregate number of securities to which transaction applies:

 

     

  3)  

Per unit price or other underlying value of transaction computed pursuant to Exchange Act Rule 0-11 (set forth the amount on which the filing fee is calculated and state how it was determined):

 

     

  4)  

Proposed maximum aggregate value of transaction:

 

     

  5)  

Total fee paid:

 

     

  Fee paid previously with preliminary materials.
  Check box if any part of the fee is offset as provided by Exchange Act Rule 0-11(a)(2) and identify the filing for which the offsetting fee was paid previously. Identify the previous filing by registration statement number, or the Form or Schedule and the date of its filing.
  1)  

Amount Previously Paid:

 

     

  2)  

Form, Schedule or Registration Statement No.:

 

     

  3)  

Filing Party:

 

     

  4)  

Date Filed:

 

     

 

 

 


LOGO

October 30, 2017

Mr. Cristiano Guerra

Head of Special Situations Research

Institutional Shareholder Services

702 King Farm Boulevard, Suite 400

Rockville, MD 20850

Dear Cristiano:

It is clear from a careful reading of ISS’s recent October 25, 2017 report (the “Report”) regarding the proxy contest between ADP and Pershing Square Capital Management, L.P. that, in reaching its voting recommendation, ADP provided non-public, inaccurate and misleading information, claims, and arguments which were relied upon by ISS. By relying on this information in formulating its recommendation, ISS diverged from its stated policy of basing its recommendations “solely upon publicly disclosed information.” Furthermore, it is clear from the Report that ISS accepted, and in its Report has publicly endorsed, certain misleading or incorrect factual assertions made by ADP management during its engagement with ISS.

ADP’s attempt to mislead ISS is a serious concern. Notably, many of ADP’s new claims and arguments are responses to critical issues raised in this proxy contest, and to questions we have publicly asked ADP that have since been privately asked by other shareholders, in each instance without a response from ADP. Rather than respond publicly, ADP appears to have misleadingly addressed these questions in private with ISS, out of the scrutiny of Pershing Square and other shareholders, in an attempt to win ISS’s support.

We do not expect ISS to have done our job and independently evaluated each of these claims, but rather to have given us an opportunity to refute these new arguments as is ISS’s customary practice. Had such an opportunity been provided, ISS might have come to a more accurate conclusion about ADP’s performance and its inadequate plans for improvement. Public discourse would have allowed ISS to properly evaluate both sides of the various arguments, and allowed shareholders and analysts to evaluate the credibility of ADP’s board and management compared with Pershing Square and its candidates.

ADP’s Misuse of the ISS Process by Providing Non-Public Information

ADP’s non-public and false and misleading information, provided privately to ISS and not otherwise made public to shareholders, analysts and the SEC, directly relates to the three key prongs of our case for change at ADP: (1) the Company has lost market share in the Enterprise

 

1


marketplace due to its lack of a competitive product offering, (2) the Company is inefficiently operated, a fact which is made clear by benchmarking ADP versus its direct competitors, and (3) ADP is unwilling to change its status quo approach and has set insufficient long-term goals.

Pershing Square was not offered the opportunity by ISS to respond to this non-public information or to rebut the false and misleading information shared by ADP with ISS that led to the ISS recommendation, nor have other shareholders or third-party ADP analysts been given a chance to review and evaluate ADP’s non-public information. As is ISS’s common practice, ISS met first with Pershing Square on October 11th, and then with ADP on October 13th. We were not made aware of or provided the opportunity to respond to ADP’s new, non-public information, claims, and arguments until the Report was issued. The Report is the first and only time that this previously non-public information has been disclosed.

In the course of all of our activist engagements over more than a decade, ISS has always offered us an opportunity to participate in an extensive and productive dialogue. This is the first time in our experience that ISS’s approach has varied from its established precedent and rules of engagement where we were not offered the opportunity to address new facts or issues raised by the subject company subsequent to our meeting with ISS in the midst of a proxy contest.

ISS’s policy is to rely only on publicly disclosed information in reaching the recommendations it circulates to its clients. Indeed, in the Report itself, in the section labeled “Engagement,” the following language appears:

ISS engages in ongoing dialogue with issuers in order to ask for additional information or clarification, but not to engage on behalf of its clients. Any draft review which may occur as part of this process is done for purposes of data verification only. All ISS recommendations are based solely upon publicly disclosed information.1

ISS’s Proxy Voting Policies and Procedures similarly state that:

ISS research and recommendations are based exclusively on public information, so any fact which an issuer would like to see reflected in our report should be publicly disclosed to all shareholders in a timely fashion; even in markets where such disclosure is not ordinarily required. Where an issuer discusses previously non-public information during an engagement with ISS, or makes a commitment with respect to future actions or behavior, that should be promptly disclosed in a Form 8-K or equivalent public filing or disclosure. In proxy contests and similar situations, we expect both management and dissidents to publicly file any materials presented to ISS. Any material non-public information that is disclosed, intentionally or unintentionally, will not be considered or included in our research reports. [Emphasis added.]

This policy makes sense as it lays the groundwork for a fair process. Among other things, ISS’s sole reliance on publicly available information ensures that each side in a proxy contest has the ability to respond to and react to the factual assertions made by the other side in full view of shareholders. As a result, neither side can attempt to gain an unfair tactical advantage by selectively disclosing certain facts and arguments only to ISS. One of the principal tenets of ISS’s Global Voting Principles is transparency, the belief that “Companies should

 

1 Report, at 2.

 

2


provide sufficient and timely information that enables shareholders to understand key issues, make informed vote decisions, and effectively engage with companies on substantive matters that impact shareholders’ long-term interests in the company.”2

ISS is an extremely influential factor in shareholder voting in proxy contests. In fact, a meaningful percentage of ADP shares are voted automatically according to the ISS recommendation, and many other ADP shareholders consider ISS recommendations in formulating their voting decision. As such, it is critical that ISS recommendations and proxy advisory reports meet ISS’s stated policies and standards so that the corporate voting mechanism is not inadvertently biased causing votes to be miscast.

In the pages that follow, we address in detail each new fact, argument and error disclosed in the Report. In light of the non-public and false and misleading information that was incorporated into your recommendation, and in accordance with longstanding ISS policy, we ask that ISS reconsider its voting recommendation and issue a Proxy Alert to its clients, both to correct factual errors in its Report which were put forth by ADP without an opportunity for public dialogue, and to ensure that ISS’s final recommendation in this contest appropriately reflects consideration of the counter-arguments Pershing Square would have made (and has made below) had it been given the opportunity to respond to the new information presented by ADP.

The ISS Report Contains Non-Public Information and Arguments, and Factual Errors, which Result in Erroneous Conclusions

Prior to making our substantial investment in ADP, we engaged in a thorough review of the Company’s public filings, conference call transcripts, and other public statements, and have continued to carefully monitor the Company’s public statements and filings during the proxy contest. Accordingly, upon reviewing the ISS Report, we were surprised to read about non-public information and new arguments that have not previously been made by ADP.

Many of these new non-public facts and arguments are intended to be responsive to questions we have publicly asked ADP, which ADP has been unwilling to answer for its shareholders. It appears, therefore, that ADP’s private and selective engagement with ISS on these points is part of a deliberate strategy to persuade ISS of certain facts and justifications in an effort to receive more favorable conclusions from ISS based upon non-public information, without giving Pershing Square, other shareholders, and research analysts the opportunity to respond to and/or ask critical follow-up questions that would demonstrate the false and misleading nature of ADP’s assertions.

In stark contrast to ADP’s approach in this regard, Pershing Square has throughout this proxy contest engaged in an open and public dialogue with ADP and other shareholders and stakeholders concerning its views, in part by asking public questions of ADP. We believe that

 

2 https://www.issgovernance.com/policy-gateway/iss-global-voting-principles/.

 

3


this open exchange is in the interest of all shareholders and is consistent with the Council of Institutional Investors’ Corporate Governance Policies.

ADP’s Supposed New Product For The Enterprise Segment

The Report notes that “the upcoming release of [ADP’s] new Vantage 2.0 platform for large clients will offer better service and help revert [ADP’s slowing growth and lost market share in the Enterprise business.]”3 The Report is the first public disclosure by ADP that the release of a new Vantage 2.0 platform is “upcoming” and will represent an improved service offering for its clients in the Enterprise segment.

We therefore question what exactly ADP said to ISS regarding its potential new product offering, which is not clear from the Report. Did the Company provide details about the product offering? Did it provide a specific timetable for the rollout of Vantage 2.0? How are analysts and other market participants able to ascertain whether this product offering has the potential (or not) to truly shift the competitive dynamics in the market? The answers to these questions are regrettably unclear as ADP has only shared this information privately with ISS.

The disclosure that ADP is close to releasing a Vantage 2.0 product offering for the Enterprise segment is material, even more so in the context of this proxy contest. One of Pershing Square’s principal critiques of ADP’s board and management has been that the Company is losing market share in the Enterprise market due in principal part to its failure to innovate and develop a best-in-class product offering. Disclosure by the Company to ISS that ADP is close to releasing an improved Enterprise product offering appears to have caused ISS to conclude that the Company is aware of and is addressing this issue, and consequently influenced ISS’s analysis and recommendation.

If indeed ADP were shortly going to release a best-in-class Enterprise product that offered a competitive level of functionality, reliability and ease of use compared with existing offerings from Workday, Ultimate Software, and Ceridian Dayforce, this would indeed be highly material information that the Company should be required to disclose in its SEC filings, or it would be in violation of Reg FD. Furthermore, under the SEC proxy rules, ADP is required to file all written material information that it directly or indirectly uses to solicit investors, which we believe include any written materials provided to ISS. Unless the Company did not file these materials in violation of the proxy rules, it appears that ADP’s representations about the release of Vantage 2.0 must, therefore, have only been made orally to ISS. How can ISS rely on ADP’s unverified oral statements about the release of a future product and its functionality and service quality?

ADP has lost thousands of Enterprise customers to competitors. If, in fact, it was about to release a best-in-class product, it would likely be in the best interests of the Company to let its clients know about the new product, its features, and functionality to help stem client defections

 

 

3 Report, at 21.

 

4


during the interim period before the product’s release. If such a product existed and could be demonstrated to shareholders, it would of course be in ADP’s interest for it to have done so during this proxy contest. In fact, ADP released a detailed presentation on September 12th and voluminous additional materials in the course of this proxy contest, and has made no such disclosures about a new Enterprise HCM product.

It is directly against ISS policy for it to consider non-public information in formulating its recommendations. In light of the materiality of the apparent Vantage 2.0 disclosure, it is particularly troubling that this non-public, unverifiable fact, apparently provided orally by ADP, was relied upon by ISS, deemed a mitigating factor to ADP’s loss of competitive position in Enterprise, and included in the formulation of ISS’s voting recommendation.

Labor Productivity and Operational Benchmarking

The Report states that:

“[I]n addressing the dissident’s criticism of employee productivity, ADP notes that employees are required to obtain and service PEO clients which generate pass-through revenues. When factoring in all revenue (including pass-throughs), ADP’s GAAP revenue/employee (averaging the starting and ending employee count in a fiscal year) was just over $215,000 in FY 2017, having grown at a 2.7 percent CAGR over the past three years. Those figures do not appear out of line with comparable figures from competitors offering payroll; ISS excluded those with a benefit focus to improve comparability.”4

ADP’s weak relative labor productivity versus competitors is a key critique Pershing Square has levied against the Company. We publicly asked the Company on October 5th why its labor productivity is materially below that of its competitors as one of our Questions of the Week. The Company has never publicly responded to our question, nor do we believe it has responded privately to shareholders asking the same question. We are not aware of any prior statements by the Company in which it suggested that PEO pass-throughs should be factored into the labor productivity calculation, or that these pass-throughs somehow provide a justification for the fact that ADP’s labor productivity meaningfully lags behind that of its competitors.

ADP’s misleading private response directly to ISS appears to have persuaded ISS to conclude that ADP’s labor productivity “figures do not appear out of line with comparable figures from competitors offering payroll.”5 ISS appears to have accepted this misleading argument as a mitigating factor in forming its recommendation. The labor productivity analysis as presented is highly misleading, as we would have explained if we had been given the opportunity to respond to this new and disingenuous argument. We believe ADP chose not to make this argument publicly because it is intellectually indefensible and dishonest.

PEO pass-throughs exist only because of the technical legal requirement that a PEO must be the “co-employer” of a client company’s employees in order to negotiate certain health-care

 

 

 

4 Report, at 21.

5 Report, at 21.

 

5


and other benefits using the group purchasing power of all of the PEO’s clients. As a co-employer, PEOs receive health insurance premiums from their clients which are “passed through,” i.e., paid to the PEO and then immediately paid out to health insurers. PEO pass-throughs are recognized in GAAP revenues and expenses, but are excluded from analysts’ and investors’ determination of a HCM business’s profitability because PEO pass-throughs are non-economic in nature. For this reason, Pershing Square has excluded pass-throughs in its operational benchmarking of ADP, a practice which both the Company and the analyst community have broadly accepted and adopted.

Had we been given an opportunity to respond to ADP’s new argument on its relative labor productivity – that the comparability of ADP’s labor productivity numbers to those of its competitors is improved when one factors in the PEO pass-through revenue – we would have explained that PEO pass-through revenues and costs are non-economic in nature and should not be included for purposes of operational benchmarking. If one properly excludes PEO pass-through revenue and costs, as we did in our analysis, ADP’s labor productivity trails that of its competitors by 28%. Alternatively, had we been given an opportunity to respond to ADP’s argument, we would have also shown that if one were to accept the Company’s proposed methodology (i.e., including PEO pass-through revenue), the underperformance becomes even more significant when properly benchmarked against the appropriate competitor set as described below.

When benchmarking ADP, one must consider that approximately 28% of ADP’s reported revenue comes from ADP’s PEO. More significantly, 21% of ADP’s reported revenue consists of non-economic pass-throughs related directly to the PEO.

 

ADP REVENUE COMPOSITION                
      2017A      % of Total  
Revenue              

    Employer Services

     $8,518        69%  

    PEO Revenue (Net)

     836        7%  

Net Operational Revenue

     $9,354        76%  

    PEO Pass-Throughs

     2,628        21%  

Operational Revenue

     $11,982        97%  

    Client Funds

     397        3%  

Reported Revenue

     $12,380        100%  

ADP is unique amongst the competitor group that ISS selected on page 21 of the Report in that it has a sizable PEO, including outsized PEO pass-throughs, while the others do not. While Paychex has a small PEO, it has a smaller amount of PEO pass-throughs due to its differential accounting and the relative size of this business compared with ADP. The other competitors presented in the table do not have PEOs at all.

 

6


ISS TABLE - ADP EMPLOYEE PRODUCTIVITY RELATIVE TO PEERS OFFERING PAYROLL  
      ISS Observation:      Pershing Square Observation:  
      Revenue / Average      PEO as % of      PEO Pass-Thoughs  
      Employee      Revenue      as % of Revenue  
            

ADP

     $215,304        28%        21%  
   
Competitor                     

    Paychex

     $231,713        11%        5%  

    Ultimate Software

     $234,745        0%        0%  

    Paycom

     $186,143        0%        0%  

    Paylocity

     $151,402        0%        0%  

    Workday

     $266,000        0%        0%  

Because GAAP PEO revenues include pass-through revenues, the labor productivity of a HCM company which owns a PEO, and whose revenues include PEO pass-throughs, will appear artificially high if analyzed on this basis – the greater the percentage of PEO pass-throughs in total revenues, the more the company’s revenue productivity will be overstated.

In its Report, ISS claims to have excluded “benefit focus[ed]” competitors from the group to “improve comparability.” Rather than improving comparability, such an approach reduces comparability. The competitors excluded are not “benefit focus[ed]” competitors but rather PEOs, and thus this methodology excludes other competitors with PEO pass-throughs, which dramatically distorts these productivity metrics. As a result, ISS compared ADP’s labor productivity, including the gross revenues from its PEO, against competitors which are not in the PEO business, or in the case of Paychex, whose PEO pass-throughs are substantially smaller as a percentage of its GAAP revenues than for ADP (5% vs. 21%).

We suspect that ISS used this approach because the revenue productivity per employee is so much greater for PEO providers than “payroll” competitors; however, that is precisely the reason why either (1) these pass-throughs should be excluded from the productivity calculations, or (2) PEO competitors must be included in the comparable set to accurately assess ADP’s relative labor productivity. ADP’s proposed framework for ISS does neither because it includes the pass-throughs to inflate productivity, but then excludes competitors which have pass-through revenues. This misleading approach led ISS to a false conclusion, namely, that ADP’s labor productivity is similar to that of its competitors.

In our analysis throughout the proxy contest, we have excluded ADP’s (and competitors’) PEO pass-throughs from the labor productivity calculation to generate an accurate, apples-to-apples comparison. Alternatively, one could do the same calculation using gross revenues including pass-throughs (i.e. the ISS methodology), but adjusting for the relative size of the PEO business within each of ADP’s competitors. We do this in the table below by including all of ADP’s competitors, including the PEO competitors.

 

7


PERSHING SQUARE - LABOR PRODUCTIVITY (INCLUDING PASS-THROUGH REVENUE)

 

            Pershing Square Observation:  
          Revenue / Average             PEO as % of              PEO Pass-Thoughs      
      Employee     Revenue      as % of Revenue  

ADP

     $215,301       28%        21%  
   

Competitor

         

Paychex

     $231,713       11%        5%  

Ultimate Software

     $235,790       0%        0%  

Paycom

     $186,166       0%        0%  

Paylocity

     $157,900       0%        0%  

Workday

     $266,001       0%        0%  

Cornerstone OnDemand

     $244,016       0%        0%  

Insperity

     $1,176,539       100%        ~76%  

TriNet

     $1,200,123       100%        79%  

Payroll / HCM Average

     $220,264       NM        NM  

PEO Average

     $1,188,331       NM        NM  

Average of ALL Competitors

     $462,281       26%        20%  
           

ADP Revenue Weighted Average (1)

     $491,163         

ADP vs. Weighted Average

     (56% )        
           

Note: Certain values have minor irreconcilable variances with ISS’s reported labor productivity.

 

(1): Revenue weighted average assigns a 28% weighting to the PEO competitors and a 72% weighting to the payroll / HCM competitors, commensurate with ADP’s revenue composition.  

Under this methodology, we show that a revenue-weighted average of the competitors – weighing the PEO competitors at 28% and the Payroll/HCM competitors at 72%, commensurate with ADP’s revenue exposures – yields a revenue-weighted average of $491,163 as compared to ADP’s revenue per employee of $215,301, suggesting ADP’s relative labor productivity is 56% below a composite of its competitors. This analysis includes all of ADP’s competitors adjusted appropriately for PEO pass-throughs, and arrives at a dramatically different and accurate conclusion regarding ADP’s relative productivity. Under either methodology (excluding PEO pass-throughs on a like-for-like basis, or including PEO pass-throughs and including PEO competitors in the competitor set), it is clear that ADP massively trails its competitors’ labor productivity, despite the Company’s enormous scale advantages.

ADP Explains Three Reasons Why Its Margins Are Lower Than Paychex’s

The Report notes that “[d]uring engagement with ISS, the board stated that several items lead its operating margin to be lower than Paychex’s,” including: (1) “Paychex does not have customers who are large businesses – this customer base has significantly lower structural margins than do small and mid-sized businesses;” (2) “ADP has higher client growth, with the incremental selling and implementation expense of adding a new client (above recurring costs) being 1 to 1.25x that client’s annual revenue;” and (3) “ADP has raised prices at 0.5 to 1 percent per year since FY 2015, slower than Paychex’s 2 to 4 percent annual price increases.”6

Once again, ADP puts forth various arguments which it has not previously argued in the public domain. Since the very beginning of this campaign, we have asked ADP to disclose sub-

  

 

6 Report, at 16.

 

8


segment margins within Employer Services, and have directly asked the Company (including on September 20th as our inaugural Question of the Week) whether or not the Company is earning comparable margins to Paychex in its comparable business segments. The Company has never publicly answered this question.

We were not provided with an opportunity to respond to this new three-part justification provided by the Company for its significant margin deficit. We note from the Report that ISS was not fully persuaded that these assertions adequately explain why “ADP’s margins fall far short of Paychex’s.”7 It does appear from the Report, however, that ISS accepted the factual validity of these arguments, but concluded that the arguments were not sufficient to account for the extent of the gap between ADP’s and Paychex’s margins. Below, we address each of these false and/or misleading responses by ADP.

With respect to argument (1), Pershing Square has never suggested comparing ADP’s overall margins to Paychex directly. We of course recognize that Paychex competes with some, but not all, of ADP’s business. That said, Paychex has both a small and medium-size customer base. Pershing Square noted on page 16 of our October 12, 2017 presentation ADP Ascending that we believe at least ~$3bn of ADP’s Employer Services revenue (~35%) is directly comparable to Paychex, and perhaps more. If ADP earned Paychex-like margins of ~41% on the directly comparable revenue, it would imply that the balance of ADP’s Employer Services business earns a ~6% margin on ~$5.5bn of revenue.

Furthermore, while Paychex does not serve large enterprises or have a sizable international business, we strongly disagree with ADP’s assertion that those business segments have “significantly lower structural margins.” Pershing Square has published an extensive body of research which demonstrates that Enterprise / international should have structurally high margins based on labor productivity benchmarking versus competitors,8 gross profit margin benchmarking,9 a careful analysis of Ultimate Software’s profitability,10 and other various public data points identifying competitors with 30% - 45% margins competing in Enterprise and/or international.11

With respect to argument (2), namely that the relative growth rate between ADP and Paychex explains the margin differential, Pershing Square largely addressed this argument on page 25 of our September 25, 2017 presentation Pershing Square’s Response to ADP’s September 12th Investor Presentation, although ADP appears to have further elaborated on this argument with ISS. In our presentation, we noted that the difference in growth rate between ADP and Paychex is not meaningful enough to explain the vast margin differential. If anything, ADP should have higher margins than Paychex due to its higher relative retention rate.

 

 

7 Report, at 22-23.

8 See page 54, ADP: The Time is Now (August 17th, 2017).

9 Id., at 56.

10 Id., at 62 – 73.

11 Id., at 74 – 75.

 

9


Ironically, the argument which ADP puts forth as a cause for ADP to have lower margins than Paychex is the exact argument we put forward to explain why ADP’s faster-growth mid-market and Enterprise competitors have lower margins today, but when appropriately adjusted for growth (and scale), demonstrate that HCM is a structurally high-margin business across all client sizes. ISS apparently accepts that “structural margins for large companies are 15 to 25 percentage points lower than for small and mid-sized companies,” a conclusion which is undermined by ADP’s own arguments about the significant upfront costs of new client wins.

With respect to argument (3), we note that this is the first time the Company has argued that differential pricing policies is an explanation for ADP’s lower margins than Paychex. We note that ADP and Paychex are both mature and scaled competitors who have competed against each other for decades. To the extent ADP had lower prices, it should have grown materially faster than Paychex over time. As noted by ADP in its September 12th presentation, ADP and Paychex have grown at an 8% and 7% CAGR respectively since FY 2011, not a material enough difference to explain any meaningful margin differential between the two.

To the extent Paychex has a different pricing strategy, it has not negatively impacted Paychex’s competitive position as both companies have similar growth rates, and there has been no observable market share shift between the two companies over longer periods of time. Paychex has enjoyed superior operating margins in the mid-to high 30% range for over a decade. They are not a recent phenomenon. In any case, our research suggests that ADP and Paychex products are similarly priced.12 Given that ADP’s RUN product is higher rated by market participants,13 to the extent Paychex has a different pricing strategy, which produces similar growth yet higher margins, ADP should adopt it as its own. Simply put, we do not think this is a credible argument to explain the vastly different structural margins between the two businesses.

ADP’s Pricing Explanation for its Gross Margin Deficit

From the Report, it appears that ADP made a new and non-public argument to ISS about its Gross Margin deficit relative to competitors, which is both inaccurate and highly misleading:

The dissident may be correct that ADP’s gross margins are below competitors, but that is not necessarily troubling. Lower gross margins can be due to a company offering its products or services more cheaply, to gain scale and build customer loyalty. They may also reflect a company’s choice to set up its business model to have a higher percentage of costs in COGS and a lower percentage in SG&A.14

 

 

 

 

12 See G2 Crowd, where both RUN and Paychex Flex score in the 77th percentile amongst user-reported pricing data measuring the “perceived cost” as reported by 1,929 user reviews. (https://www.g2crowd.com/products/adp-run/pricing and https://www.g2crowd.com/products/paychex-flex/pricing)

13 Supported by our broad market research but also easily observable on such websites as G2 Crowd where RUN by ADP scores a 4.5/5.0 stars while Paychex Flex scores a 2.9/5.0 stars.

14 Report, at 22.

 

10


It is incorrect that ADP’s lower gross margins are due to its decision to offer products and services at a lower price point in order to gain scale and build customer loyalty (e.g., like Amazon). In fact, ADP has taken the exact opposite approach. ADP is the mature and scaled industry incumbent with leading, but declining, market share. The above claim would imply that competitors are “higher-priced” yet still gaining share, which could only explain that ADP’s product and service offering must be highly inferior.

It is empirically false that ADP’s products are broadly offered at a discount to its competitors. This purported explanation for the gross margin differential between ADP (58% gross margins in ES) and its competitors (74% recurring gross margins) would be true only if ADP’s products were priced at a ~20% or more discount to competitors. Such a pricing discount to “gain scale” or market share should have driven significant market share gains, yet ADP is losing share as shown in our analysis and acknowledged by ISS.

If anything, the market share gains of ADP’s competitors would imply the opposite, that competitors are offering products at lower price points to gain share from ADP, and that ADP’s competitors should therefore have lower gross margins. Based on our research, competitors are not pricing at a discount, but rather are gaining share versus ADP because they have better technology and service offerings. In fact, competitors and industry observers note that pricing is a very small consideration in selecting vendors as we have previously explained.15 ADP’s gross margins are lower than competitors’ not due to price, but because the Company is extremely inefficient as we have explained in detail.

Moreover, ADP’s lower margins are demonstrably not the result of the Company’s “choice” to allocate a higher percentage of costs to COGS and a lower percentage to SG&A. Pershing Square’s analysis adjusted ADP’s and its competitors’ gross margins to improve their comparability in our benchmarking analysis. ADP’s GAAP actual gross margins (as reported in its financial statements) are 41%. We adjusted ADP’s gross margins upward so that they can be appropriately compared to competitors. After these comparability adjustments, ADP’s adjusted gross margins are ~60% (and ~58% in Employer Services), which still significantly trail competitors’ 74% recurring gross margins. In all instances, we attempted to isolate gross profit and other cost line items below gross profit in order to enable a proper comparison across all competitors.16 While it is possible that companies may classify costs differently, we have made appropriate adjustments to account for this issue. Moreover, this would only provide an explanation for ADP’s lower margins if its operating margins were in fact the same as

 

 

 

15 See page 26 – 27, ADP Ascending: Detailed Supporting Materials (October 11, 2017). Note that amongst new Ultimate Software clients, when asked the question “What factors are allowing Ultimate to win versus incumbent providers?” only 16% identified “cost savings” as an influential factor. It ranked lowest amongst disclosed reasons for switching from an incumbent vendor to Ultimate Software.

16 As noted in our original presentation ADP: The Time is Now, gross profit margins were adjusted to improve comparability amongst competitors and ADP. For ADP specifically, gross margins were adjusted to exclude float income (-$397m) but add-back Systems Development and Programming Costs (+$628m) and Depreciation and Amortization (+$226m). In all instances, gross profit margins were presented as a percentage of adjusted net operational revenue, excluding float income and reporting the PEO net of pass-through costs.

 

11


competitors. In this case, ADP’s gross profit and operating margins massively trail those of its competitors (e.g. Paychex).

ADP Claims Our Plan Includes “Drastic” Headcount Reductions And Creates “Risk”

The Report notes that “the board argues that the dissident has not been forthcoming about the fact that its plans to raise margins would involve a drastic reduction in the number of ADP employees. Such a large headcount reduction, the board states, could disrupt client relationships and customer service, which would put margin improvements at risk.”17

The Report then further reflects on this assertion as part of its determination for the question “Is Change Warranted,” noting “[a]lthough the dissident has not explicitly called for headcount reductions, it appears unlikely that substantial margin improvements, such as those promoted by the dissident, could be achieved without such cuts.”18 ISS appears to have heavily weighted this line of reasoning and accepted it as support for the Company’s claim that our plan is not “devoid of risk.” Pershing Square notes that ADP had not raised this argument directly prior to our meeting with ISS, nor was it discussed in our meeting with ISS.

First, the claim that we have “not been forthcoming” about the need for “drastic” headcount reductions is due to the fact that this claim is not true. The HCM industry is a high-growth industry – ADP’s own plan calls for top-line growth of 7% to 9%. The benefit of growth is that it provides an opportunity for ADP to “grow into” its labor productivity, as Paychex has done over the past six years. In FY 2009, Paychex had 12,500 employees and generated $2bn of revenue. In FY 2017, Paychex employee count had risen to 13,700 (+10%) and its revenue has grown to $3.1bn (+54%). Paychex’s labor productivity has expanded from $161k per employee in FY 2009 to $226k today, despite its headcount growing over the period as revenue has grown far in excess of headcount.

Similarly, if ADP were to achieve the growth rates we believe it should, it would grow into its headcount over the coming years. Furthermore, ADP has a relatively high associate turnover – we understand that ADP’s associate attrition rate is >10%. As such, ADP needs to hire ~6,000 associates every year simply to keep headcount flat. ADP could drastically accelerate labor productivity by simply hiring fewer new associates and allowing the Company’s natural attrition to run its course. If ADP simply held its net headcount flat, it would achieve competitor-level labor productivity within ~five years (consistent with Pershing Square’s FY 2022 margin target) based on the Company’s stated 7% to 9% revenue growth goals.

 

 

17 Report, at 17.

18 Id. at 25.

 

12


ILLUSTRATIVE ADP LABOR PRODUCTIVITY

 

      FY 2017      Year 1      Year 2      Year 3      Year 4      Year 5  
                     

Net Operational Revenue

     $9,354        $10,102         $10,911         $11,783         $12,726         $13,744   

Growth (%)

        8.0%         8.0%         8.0%         8.0%         8.0%   
   

Starting Employees

        58,000         58,200         58,380         58,542         58,688   

(-) Natural Attrition (~10%)

        (5,800)        (5,820)        (5,838)        (5,854)        (5,869)  

(+) New Hires

        6,000         6,000         6,000         6,000         6,000   

Ending Employees

     58,000        58,200         58,380         58,542         58,688         58,819   

Growth (%)

        0.3%         0.3%         0.3%         0.2%         0.2%   
                     

Labor Productivity

     $161        $174         $187         $201         $217         $234   

Alternatively, if ADP hired ~4,000 new employees a year (vs. 6,000 shown above), ADP’s headcount would decrease modestly, and ADP would achieve competitor-level labor productivity levels within three years.

Notably, CDK has managed to achieve enormous margin gains while keeping headcount relatively flat – which further supports our views on the opportunity to improve labor productivity and margins without drastic headcount reductions.

 

CDK LABOR PRODUCTIVITY

                        
     

FY 2014

   FY 2015     FY 2016     FY 2017  
             

Revenue

   $1,977      $2,064       $2,115       $2,220  

Reported Growth (%)

        4.4%       2.5%       5.0%  

Organic Revenue Growth (%)

     7.5%       6.8%       5.8%  
             

Employees

   ~9,000      8,900       8,700       8,900  

Growth (%)

        (1.1%     (2.2%     2.3%  
             

Labor Productivity

   $220      $232       $243       $249  

Growth (%)

        5.6%       4.8%       2.6%  
             
             

Adj. EBITA Margin (%)

   16%      18%       22%       26%  

ADP’s claim that our plan includes “drastic” headcount reductions is a disingenuous argument put forth by ADP management to scare employees and shareholders into supporting the status quo and to cause others to reject our plan as “risky.” This argument is not supported by the facts and, therefore, should not have been considered by ISS in formulating its recommendation.

ADP’s Purported Plans for 500 to 600 Basis Points of Margin Expansion

ADP has recently made a series of public statements that its plan will lead to net operational margin improvements of “500 to 600 basis points” by FY 2020. This is a new and inaccurate disclosure that we believe is demonstrably false, and is intended to mislead shareholders and others about the lack of progress implied by ADP’s plan. Although ADP did not emphasize its so-called plans for 500 to 600 basis points of margin expansion at the start of this proxy contest, after recognizing significant shareholder support for our view that ADP is

 

13


inefficiently operated, the Company has focused almost exclusively in recent weeks on its purported claim that its plan will deliver “500 basis points” of net operational margin expansion.

ADP included only a small reference, as a sub-header to one slide, to its purported 500 basis points of net operational margin expansion over the next three years in its investor presentation released on September 12th. ADP otherwise had said nothing about this 500 basis points until October 18th, at which point, the Company made its so-called plans for 500 basis points of margin improvement a centerpiece of its argument that shareholders should support the current board and management. We believe that ADP has done this to mislead investors about the lack of progress in its plan, and that this pivot in message is in direct response to shareholder support for our views.

Pershing Square asked ADP to provide support for this 500 basis points figure in our Question of the Week on October 19th. We are aware that numerous shareholders and sell-side analysts have contacted ADP’s Investor Relations team and management to clarify this figure to no avail. Again, today, we have publicly asked ADP either to support its “500 basis points” claim or retract it. ADP should be required by ISS to immediately provide support for its 500 to 600 basis point claim. Such a response should be immediately made public so that all shareholders and ISS can determine its validity.

ISS seems to have accepted ADP’s 500 to 600 basis points claim:

The company expects its operational EBIT margin to improve by 500 to 600 basis points by FY 2020, with two-thirds of that gain coming from operating leverage, and the rest resulting from initiatives such as migrating customers away from older IT platforms and reducing the number of corporate offices. These plans suggest that the board and management may not be content with the current situation.” [Emphasis added.]19

Unfortunately, and inexcusably, in all of ADP’s recent public statements, ADP has failed to make clear to shareholders and analysts that its plan contemplates overall margin improvement of only 100 to 200 basis points, and net operational margin improvement of just ~300 basis points.

Pershing Square addressed this misleading 500 basis points claim on pages 21 and 22 of our September 27th presentation (which was further included at pages 62 and 63 in the detailed supplement to our October 11th presentation, ADP Ascending). Net operational margin is calculated as (EBIT – client funds interest – corporate extended income net) / (total revenue – PEO pass-through revenue – client funds interest).20 This calculation can be determined with just five figures, which are all reasonably estimable from ADP’s September 12th plan. When

 

 

 

19 Id. at 23.

20 Consistent with ADP’s financial reporting, both client funds interest and corporate extended investment income are deducted from EBIT in the numerator, while the interest on client funds deducted from revenue in the denominator does not include the effects of the extended investment strategy.

 

14


one does the math, one learns that ADP’s announced plan implies not 500 to 600 basis points of improvement, but rather just ~300 basis points of net operational margin expansion.

ADP’s current net operational margins are 21.6%. The Company’s current plan provides revenue guidance of $15-$16bn by FY 2020 ($15.5bn at the midpoint) and margin guidance of 21-22% (implying ~$3.3bn of EBIT at the midpoint). Given other known variables, including the Company’s guidance for its HRBPO business (including its “PEO”) to grow at a 12 – 14% rate, it is effectively a mathematical impossibility for the Company’s current plan to produce 500 basis points of net operational margins, but only deliver an actual margin increase of 100-200 basis points. By our calculation, ADP’s plan delivers net operational margins of ~24.5%, or just ~300 basis points of net operational margin expansion. Our detailed deconstruction of ADP’s announced September 12th plan shows this ~300 basis points of operational margin expansion (of which ~2/3rd is from Employer Services), which can be found on page 63 in the detailed supplement of our October 11th presentation, ADP Ascending.

Rather than clarifying this claim for shareholders, ADP has continued to prominently feature this misleading claim in an attempt to win support from shareholders and to help persuade influential observers such as the proxy advisory firms. We note that Glass Lewis correctly saw through ADP’s misleading claim:

Upon closer scrutiny, ADP’s three-year plan calls for only approximately 300 basis points in net operating margin expansion, and only 220 basis points of margin expansion in Employer Services, which is consistent with ADP’s historical performance and long-term plan of delivering 50 to 100 basis points per year. Upon review, we find the Company’s plan to be underwhelming in this regard, as does Pershing Square, other ADP shareholders we spoke with and certain analysts who follow the Company.21

In its Report, ISS relied on ADP’s unsupported claim of 500 to 600 basis points of future margin improvement and concluded that “the board and management may not be content with the current situation.” ISS appears to have accepted this false margin claim which allowed ISS to discount our claims that ADP is complacent with the status quo. That is among other reasons why we and other shareholders believe real change in the form of partial board turnover is needed.

ISS’s Recommendation to Facilitate My Election to ADP’s Board Is Unprecedented

We note that ISS recommended my election to the board, but did so in an unprecedented manner. Namely, ISS recommended that shareholders who wished to support my candidacy according to your recommendation do so by withholding their votes for Eric Fast on management’s white card. ISS did not recommend that shareholders submit their votes on the GOLD card, the only place where Pershing Square’s nominees appear.

 

 

21 Glass Lewis report, October 23rd, at 23.

 

15


A withhold vote for Mr. Fast is not a vote for me. I will be elected to the board only if the votes I receive on the GOLD card exceed the votes Mr. Fast receives on the white card. In other words, ISS is apparently attempting, in ISS’s own words, to “facilitate” my election by relying on the votes of shareholders who reject ISS’s recommendation, and instead vote on the GOLD card.

Our proxy advisor has reviewed all of ISS’s historical recommendations and cannot find one precedent where a similar approach was used by ISS to achieve this recommended outcome.

Alternatively, ISS could simply have recommended that shareholders vote for me on the GOLD card, and withhold votes for our other candidates. Not only is this the simpler, customary, and more straightforward approach, it is more likely to achieve ISS’s recommended outcome.

In the Report, you note that the use of a universal proxy card would have been optimal for shareholders:

For shareholders, the optimal solution would likely have been – as the dissident suggested – for the board to adopt a universal proxy card for this contest, which would have allowed them to cast unfettered votes for the candidates of their choice regardless of the slates. Given the timing of the dissident’s request, however, it is difficult to assert that the board’s response to this was unreasonable in this case.22

We met with ADP’s full board on September 5th to present our views on the Company and to outline the value our directors would bring to the board. ADP subsequently rejected our candidates and proposals on September 7th. On September 15th, eight days after our candidates and proposals were rejected by ADP, we proposed to the Company that it use a universal proxy card in the ongoing proxy contest. With more than 55 days to go to the election, there was no reason the parties could not have worked together to utilize a universal proxy card.

While ADP publicly claimed that using a universal proxy card would have “confused” shareholders, nothing could be further from the truth. The current U.S. system of two proxy cards with different slates is inherently confusing to shareholders. A universal proxy card is more easily understood by all investors, as it is the approach used for all other political or other elections of which we are aware. Many investors we have spoken with have expressed serious concerns with ISS’s proposed mechanism to “facilitate” my election to the board. These investors note that such an approach is extremely confusing and will disadvantage my candidacy while advantaging the incumbent directors when compared with the more straightforward and customary approach of directly voting for me on the dissident proxy card.

ADP’s rejection of the universal proxy card had nothing whatsoever to do with timing or any concerns about confusing shareholders. Rather, ADP viewed a universal proxy card as a tactical disadvantage and therefore rejected it. It is disappointing that ISS accepted the board’s

 

 

22 Report, at 26.

 

16


self-interested justification for the rejection of a universal proxy card in light of the benefits to all shareholders which would have resulted.

Furthermore, you did not note in your Report, but could have noted, that shareholders of record or shareholders who hold legal proxies who attend the meeting or send an authorized representative on their behalf could use the Company’s ballot at the meeting to accomplish what could have been accomplished if a universal proxy card had been used.

Conclusion

It is regrettable, and unusual in our experience, that Pershing Square was not provided with an opportunity to address ADP’s non-public and false and misleading assertions with ISS in advance of the issuance of the Report and its voting recommendation. ISS’s own policies and procedures favor the kind of engagement that was lacking here:

 

    “Our goals are to facilitate productive and informative dialogue….”
    “ISS’ research teams interact regularly with company representatives, institutional shareholders, shareholder proponents and other parties in order to gain deeper insight into many issues and to check material facts relevant to our research.”
    “As a research organization, we encourage constructive dialogue among stakeholders on critical governance issues to ensure a fuller understanding of the facts and circumstances, which in turn will enrich and inform our analysis and recommendations.”23

We had a productive meeting with ISS representatives on October 11th. Thereafter, we offered to make ourselves available as needed for any questions and further discussion. We did not hear from ISS again, including after ISS’s meeting with ADP on October 13th.

Had we had the opportunity to discuss ADP’s non-public information, claims, and arguments that it shared with ISS before the issuance of the Report, ISS might have reached a materially more favorable conclusion for Pershing Square, and supported all three of our nominees on the GOLD card. At a minimum, ISS might not have made the unprecedented voting recommendation it did. ISS’s support for nine of the ten incumbent directors clearly seems to have been based in significant part on the belief – which in turn was supported by non-public information and erroneous calculations and conclusions based on misinformation provided by ADP – that the current board is aware of the issues Pershing Square has highlighted and is already proactively and effectively taking steps to address them.

We understand that ISS strives to be as accurate as possible in its research and its publications. As ISS’s policies note: “Significant factual errors are rare in our reports but ISS always strives to be as accurate as possible. If you believe our report contains an error, please notify us immediately. If we agree that there is a material error that should be brought to our

 

  

 

23 https://www.issgovernance.com/contact/faqs-engagement-on-proxy-research/

 

17


clients’ attention, we will promptly issue an “Alert” and revised report to all our relevant clients.”24

We ask therefore that you carefully reconsider your Report and its conclusions in light of the issues that we have identified above and promptly issue an “Alert” report with any updated conclusions and voting recommendations. To the extent that ADP has responses to our responses, we are available to promptly address any rebuttals that it raises. In reconsidering your recommendation in light of the issues we have raised in this letter, we also ask that you reconsider the voting mechanism for my election and recommend the more traditional and customary approach of voting for the preferred candidate on the dissident proxy card to better facilitate my election.

We believe that it is critically important that ISS live up to its own published standards of doing business that are designed to facilitate its ability to arrive at an independent conclusion following a public exchange of facts and arguments by both sides. The non-public nature of the information and arguments ISS appears to have relied upon in preparing its Report in this proxy contest, along with the unprecedented nature of its recommendation, raise serious concerns that this particular engagement did not live up to those standards.

We note that the election is one week from tomorrow. It is critical that ISS immediately issue a corrective report and a proxy advisory so that investors, particularly those using ISS’s automated systems, have the ability to make changes to their votes in advance of the election.

Thank you for your consideration of these critically important issues.

Sincerely,

/s/ William A. Ackman

William A. Ackman

Copy:

Gary Retelny, President and CEO of Institutional Shareholder Services

 

 

24 https://www.issgovernance.com/contact/faqs-engagement-on-proxy-research/

 

18