Description

We are recommending a short position in NCR.

After noting substantial, widespread, and large (as % of total) insider selling (at levels well below current market price), multiple acquisitions, and extremely weak FCF/NI conversion (even after adjusting for pension contributions), we examined the underlying drivers of NCR’s recent strong growth in the company’s all-important NPOI (non- pension operating income) and non-GAAP EPS. Our objective was to divorce the economic reality from any optics driven by acquisitions and accounting policies. Further piquing our interest was that Chairman & CEO, William Nuti, was involved  in various accounting and financial control irregularities while supposedly “cleaning up” Symbol’s prior accounting problems. At Symbol, Mr. Nuti faked a turnaround, produced a few good quarters, got stock to jump, dumped  all his stock, and then revealed that continued problems with internal problems meant those results could not be relied upon.  He then was forced to resign.

Several members have done a nice job of laying out NCR’s various businesses. As a result, this analysis pre-supposes some degree of familiarity and is confined to questions involving NCR’s potential use of acquisitions, liberal accounting, and one-off asset sales to distort the underlying operating performance of the company. We first catalogue some of the devices that NCR has employed to show an upward march in Revenue and NPOI between ’11 and ’13. We then combine the effects of some of the distortions with the contributions from the two large (Radiant and Retalix Acquisitions) and several bolt on acquisitions made over this period. This analysis shows that NCR’s organic business performance is much weaker than commonly perceived. This divergence between NCR’s relatively weak organic business trends and its ambitious financial targets gave birth to the Digital Insight acquisition. The inflated price paid for Digital Insight (16x EBITDA and, more importantly, a 60% premium to price paid by Private Equity just 3 months earlier) truly reveals just how desperate NCR was to close a sizeable transaction (before lapping Retalix) in order to continue to obfuscate underlying operational results. Impressively, this was the 3^rd large acquisition in 3 years, all closing in the 2^nd half of the year. We do not believe this is a coincidence.

Non-Sustainable Drivers of NCR’s seemingly impressive Results

 NCR’s Revenue, NPOI, and non-GAAP EPS climbed by 8.3%, 22%, and 21%  from FY ’11 to FY ’12. Revenue, NPOI, and non-GAAP EPS were +9%, +21.6%, and +12% from Q3 ’12 to Q3 ’13. NPOI margins of 9.1% in ’11 are expected be 11.6% in ’13. These results are more or less consistent with NCR Projections (through ’16) of 7% Revenue CAGR and NPOI margin expansion to 14%-15%. In digging into specific variances behind these impressive results, several red flags emerged.

(1) Commingling Current Asset Sales with Future Product Sales

 In two instances, NCR has entered into arrangements whereby it sells a non-core operating business in return for both cash and a commitment by the acquiring entity to purchase goods or services from NCR for a predetermined amount. The distortion here is that the commitment extracted from the acquiring entity unduly flatters the results of the remaining business. The revenues and operating income stemming from such arrangements should be considered "one-off" asset sales rather than recurring revenues and should be removed from the results of existing operations. Clearly, by converting asset sales into recurring revenue, the company is hoping that investors mistakenly capitalize the latter, inflating the company's valuation.

 A. Sale of NCR Brazil (49%) to Banco Bradesco in 2011

 In 2011, NCR sold its 49% stake in NCR Brazil to Scopus (an entity owned and controlled by Banco Bradesco) for $43 million and a commitment by Bradesco to purchase 30k ATM's over 5 years.  Revenues from this contract thus far have been $35 million in 2011 (Q4 only), $145 million in '12, and $101 million through Q3 ’13 (+$6 million YoY).

 B. Redbox/Outerwall Kiosk Sale

 NCR entered into a similar arrangement with Redbox in 2012 whereby it sold its Blockbuster Video Kiosks for both $25 million in cash and a guarantee from Coinstar to buy $20 million in gross margin worth of goods and services by '15. Only the cash portion was recorded as consideration for the sale, even though the guarantee from Coinstar was clearly consideration that was received as a function of the asset sale. This will enable NCR to show $20 million in gross margin that I believe is more properly $20 million related to asset salees. 

 (2) Rapid increase in Cap. Ex/Sales suggests potentially aggressive capitalization

 While Revenues were +8.3% YoY from '11 to '12, capital expenditures were +30% YoY, from $123 million to $160 million (+$37 million). While the company attributed this to Radiant acquisition, this makes little sense as Radiant's historical capital expenditures were ~ $6.5 million per annum and this $6.5 million increase would only have had an impact for the 8 months in '12, so $4 million. 

The company did not disclose any ambitious growth cap. ex. If we assume that cap. ex had grown at the same rate as revenue (+8.3% YoY), then 2012 cap. ex would have been $133 million. So if we assume that the gap between the $160 in reported cap. ex and $133 in actual cap. ex is more aggressive classification of expenses into capital expenditures, the NPOI increase from this maneuver would be $27 million.

 2013 is even more aggressive.  9 months '13 cap. ex is +40% YoY (from $111 million to $155 million) while revenues are + 8.9% YoY. Assuming more aggressive reclassification of op. ex into cap. ex, this imbalance (between growth in Revenue and growth in cap. Ex) contributed $34 million of the $88 million positive variance in NPOI. 

(3) Post-Employment Expense and Post-Retirement Expense

 While NCR has done a commendable job of shrinking its Unfunded Pension, it also has other non-pension Post-Employment and Post-Retirement Benefit Plans of $246 million that are separate from the Defined Benefit Pension Plan. There are no assets set aside to pay these benefits; they are "pay as you go" and they run through the P/L (and are included in calculating NPOI). As with pension expense, there is a service component (reflecting benefits earned by current workers), an interest component (on previously accrued obligations), and other actuarial gains/losses that might result from changes in assumptions underlying the plan. For analytical purposes (as opposed to accounting purposes), only the service component should be treated as an operating expense with all other charges below the line as financing charges.

 NCR elects to treat the entire charge as an operating charge. While this seems conservative at first glance, it has distorted economic reality. How? Over the last 3 years (2010, 2011, 2012) the non-operating, non-service expense/(benefit) of/(from) Post Employment and Post Retirement Expense was $17 million, $8 million, and ($1 million). This decline was due to partial funding of the obligation and actuarial amendments to plan, rather than an actual improvement in the underlying operating business. The distortion is that by keeping it above the line, NPOI improves by $9 million in 2011 (from $17 million expense to $8 million expense) and then another $9 million in 2012 (from $8 million expense to $1 million benefit) and NPOI margins display a steady increase that is not reflective of any underlying improvements in the business.

 In the Q1 '13, this was particularly influential because of a shift in accounting procedure (FAS 112) that enabled NCR to record a $13 million curtailment gain in its Post-Employment Expense. This curtailment drove an $11 million benefit from Post-Employment Expense in Q1 '13 vs. a $2 million expense in Q1 '12, thereby adding $13 million in Year over NPOI. In Q2, without the curtailment gain, Post Employment Expense was flat with the previous year.

 4) Gains from Real Estate Sales included in Recurring Earnings 

 While NCR does not hesitate to add-back to NPOI any and all non-recurring charges, it does not similarly subtract gains from non-recurring asset sales from its NPOI. Besides just overstating NPOI and NPOI margins, increases in gains from asset sales over recent years have distorted operational reality by giving the appearance of annual NPOI increases due to margin expansion.

 For example, Gains on Sale of PPE jumped from $5 million in '11 to $10 million in '12, which enabled NPOI to increase by $5 million (and also facilitate NCR's margin goals).

 In the 9 months through '13, gains on sale of PPE have been $14 million vs. $6 million during the comparable time period in '12. This increase in gain on sale has enabled NCR to record an $8 million YoY increase in NPOI (and further boost NCR's margins).

 (5) Vendor Rebate in Q2 '13

 In Q2 '13, NCR Financial Services OPM climbed from 10.9% to a record 12.1% despite a YoY decline in sales. The company cited lower expenses due to a "one-time" reimbursement from a supplier. The one-time reimbursement represented a "netting of cost that incurred over the last 12-18 months".

 NCR would not quantify it. If we assume that margins would have been flat (a fair, if not generous assumption given the lower margin International Mix in the quarter). The supplier re-imbursement would have represented a one-time credit of ~ $10 million in Q2.

 (6) Warranty Expense

 Throughout 2011 and 2012, NCR's warranty accrual expense as % of Revenue was between 75 and 85 bps every quarter and roughly matched the corresponding warranty settlement (cash outflow). Yet in 1H '13, warranty accrual expense as % of Revenue was 58 bps, down 21 bps from 1H '12. The accrual amount was only 85% of the settlement amount. Had warranty accrual as % of Revenues remained constant, 1H' 13 Gross Margins would have been 22 bps lower and 1H '13 NPOI would have been $6.66 million lower.

 (7) DSO Explosion

 DSO’s +20% YoY in Q3 ’13 from 58 days to 69 days.  DSO’s +12% YoY in Q3 ’12 from 52 days to 58 days. DSO’s +32% from Q3 ’11 to Q3 ’13 (from 52 days to 69 days). Management has not given a strategic explanation for the increase. The CFO noted on Q3 '13 call that they "collect more of their receivables as % of Revenue in last quarter of every year", but this does not address the YoY increase.

 (8) Retalix Post Announcement/Pre-Close Period

 When NCR announced its $650 million acquisition of Retalix on 11/28/12, it forecast Retalix '13 EBITDA of $38.2 million. It cited Retalix financials (and Retalix was a public company with long record of financial data) as being ~ 44% GM and ~ 10% OPM. With consensus expectations for Retalix '13 Revenues of $280-$300 million and $30 million in EBIT (trailing 12 month D&A was ~$7 million) NCR's forecast seemed reasonable. If we added the $5-$10 million in synergies that NCR expected, we would get Retalix contribution to NCR's NPOI of ~$35-$40 million for a full year in '13 (11.5%-13% OPM), or $10-$11 million per quarter.

The deal closed on 2/5/13, 69 days after it was announced. NCR included Retalix in its financials from 2/6/13 onward, or 54 out of the 90 days in Q1 '13. During those 54 days, Retalix reported $50 million in Revenue and $9 million in NPOI (18% adjusted OPM). This 54 day contribution would have worked out to $83 million in Revenue and $15 million in operating income for the full quarter. Such a number would have meant Retalix Revenues of $334 million for FY '13 and Retalix NPOI of $60 million for FY '13 vs. the $280 - $300 million in Revenues and $40 million in NPOI expected assuming the top end of NCR's synergy target. 

In Q2 '13, Retalix produced blow-out numbers of $80 million in Revenues and $16 million in NPOI (20% OPM and 2x what Retalix historical OPM had been). Again, the result, if annualized, would imply Operating Income nearly double analysts targets and more than 60% greater than a forecast assuming the high end of the range of NCR synergies.

What explains the divergence between Retalix pre and post-acquisition?

While an improvement in Retalix business could have occurred, such a wide disparity from analyst’s pre-merger forecasts raises some other possibilities. For example, we know that Retalix recorded only $20 million in Revenue in the 36 days from 12/31/12 to 2/5/13 (before closing). This $20 million implies a quarterly revenue run-rate of $50 million and annual revenue run-rate of $200 million. This is in stark contrast to the quarterly revenue run-rate of $83 million ($330 million annual) implied by the 54 days from 2/6/13 to 3/31/13 (after the deal closed). It is also well below the $75 million quarterly run-rate ($300 million annual) Retalix analysts had forecast for ’13.

Assuming the 33 day period from 11/25/12-12/31/12 experienced a similar revenue trend to the 36 day period from 12/31/12 to 2/5/13, total revenue recognized during the pre-close period would have been $38 million. This $38 million would have been ~$19 million less than the $58 million a $300 million annual run-rate would imply; leaving $19 million “extra” revenues available for post-close period. Retalix Revenue during the 233 days it has been owned by NCR has been $212 million, $20 million higher than would have been expected using revenue run-rate of $300 million. At 45% Incremental margins (assuming GM ~ Incremental Margin), that extra $20 million would translate into $9 million inextra NPOI.

So we note the possibility that Retalix deferred Revenue during the pre-close period and this had a leveraged impact on NPOI. If so, this “excess” revenue should soon be exhausted,  reversing any beneficial impacts from Q1’13- Q3 ’13. In this vein, it is interesting that both the CEO and CFO of Retalix  resigned in September- just 8 months after close. This was despite the fact that the merger documents only entitled the Retalix CEO to a bonus 18 months after the close

 NCR’s Adjusted, Organic Business Performance

 When we include both the effects of the accounting maneuvers and the contribution from acquisitions, we get the following results for 2011/2012 and 9 months 2012/ 9 months  2013.

2011/2012

Revenue Growth 2011/2012 (Organic vs. Acquired)

NPOI Growth 2011/2012 (Organic & Undistorted vs. Acquired & Distorted)

Revenue Growth 9 Months  2012/2013 (Organic vs. Acquired)

NPOI Growth 9 Months 2012/2013 (Organic & Undistorted vs. Acquired & Distorted)

So the above analysis shows how the underlying economic performance of NCR is performing much worse than is commonly expected with LSD organic revenue growth and margin compression, leading to organic NPOI declines. this does not include other potential areas of manipulation such as rise in DSO's or other purchase accounting and cash integration charges (which have also increased YoY). These numbers should be compared to the lofty targets NCR set out at its’13 Investor Day of 9% Revenue CAGR, 13% NPOI margin (from 10.3% in ’12), and 15%-20% non-GAAP EPS (to $3.90 by midpoint ’15)

Despite NCR’s flattish to negative organic growth in NPOI through Q3 ‘13, the street was still modeling 15.5% NPOI growth for ’14 in late ’13. Failure to execute a substantial acquisition in late ’13 would have forced the company to materially bring down consensus top line and NPOI estimates. We believe there were very few candidates that could be both rationalized to investors and were large enough to obfuscate the company’s weak organic trends. It is these pressures that gave birth to the $1.65 billion Digital Insight acquisition. The media and investment community have (rightly) shaken their heads at the price (65% premium to price paid by Private Equity just 3 months earlier) but have (wrongly) concluded the source of Thoma Bravo’s windfall was Intuit’s sloppy sales auction. In reality, the astronomical return is more likely a function of NCR’s desperation and the paucity of targets that could fit its needs and, importantly, be found in such a short time frame.  Given the circumstances surrounding the investment, we believe Digital Insight will prove to be a negative NPV commitment.

Valuation/Downside

The bull case points to ~ $3.75 in EPS by ’15  (however achieved) , puts a 15x forward P/E multiple on this, and gets $56 stock by end of ’14, for ~ 50% upside. At a minimum, the current price is only 10X ’15 EPS which is extremely cheap in the current market climate. We disagree entirely with this logic for the following reasons,

(1)     Using a P/E multiple to value a company with 4X of Leverage where cap. ex > depreciation, tax rate in mid-20%, and history of weak FCF conversion seems absurd and strikes us as "bull market math."

(2)     If we are right (that acquisitions and aggressive accounting have driven 100% of non-GAAP EPS growth) the company’s growth trajectory is unsustainable as the 4X of leverage should preclude the company from future acquisitions large enough to move the needle, and accounting and non-operational levers are closed to maxed out. This latter point is important considering that the increased leverage should bring greater investor scrutiny on FCF.

We value NCR on EV/ (EBITDAP less cap.ex). Pro-forma for Digital Insight and Aleric acquisitions, NCR’s Net Debt climbs to $3.6 billion from $1.9 billion at Q3 ’13 (using the $350 unrestricted cash balance that management cites in Nov. Investor presentation). NCR has close to $150 in Enviro. Obligations and ~ $100  in unfunded pension, so TEV is ~ $10.34 billion ($3.6 billion Net Debt +$200 enviro & pension + (170 million shares @ $37)).

Consensus ’14 NPOI is $820 million. Bulls will note that this $820 million is the same as before the Digital Insight Acquisition (numbers are stale) despite the company claiming that Digital Insight will add ~ $100 in incremental NPOI (offset by the incremental interest expense of $80 million). So pro-forma for the Digital Insight Acquisition, NPOI should be ~ $920 million. Yet we do not believe the base business would have seen NPOI increase from $710 in ’13 to $820 in ’14. Indeed, we believe that Digital Insight acquisition was likely done to make $820 million number achievable. Assuming base business is flat ($710 million), and Digital Insight adds $100 million in NPOI, we arrive at $810 million in NPOI. Adding $145 in D&A and subtracting $225 in expected cap. ex gives pre-tax UFCF of $730 million. At a valuation of 10x pre-tax UFCF, we drive a price target of $21, or 43% downside from current levels.

Catalyst