StarTek's First Combined Entity Quarter: Many Hidden Bright Spots

|
About: StarTek, Inc. (SRT), Includes: AMZN, TLPFF, TLPFY
by: Omar A. Samalot
Summary

Truly global, diverse revenue with not one client reaching the 10% of revenue threshold!

On a constant currency basis, revenue declined $11.7m year over year due to the well-reported weakness in U.S. wireless related business.

Management indicated during the earnings call that "most of the decline in telecom, particularly in the U.S. side, has already happened through the system."

Integration ahead of schedule with $12.5m in annualized cost savings already implemented; benefits from these savings are expected to start showing up as early as the current December 2018 quarterly results.

As per information contained in the 10Q filing, the Non-GAAP Net Loss for the full quarter, including the 19-day stub period of Startek-only results, was ($0.1m); essentially breakeven.

Investment Thesis Follow-up

The investment thesis described on my last Top Idea article remains intact with Q3'18 StarTek/Aegis combined results confirming a number of points:

  • Zero merger overlap resulting in the integration progressing ahead of schedule and in the implementation $12.5m in annualized cost savings; all of this within the first 4 months after combination.
  • Elimination of revenue concentration risk. Not one client reached the 10% of revenue threshold as per the Q3'18 Form 10Q filed on 11/09/18. StarTek (SRT) now serves over 250 clients across 13 countries in over 40 languages. Source: Sept. 2018 Investor Presentation
  • Despite the weakness in the wireless/telecom related business, the company experienced strong revenue growth from non-telecom verticals including sequential growth from every top ten client; estimated, combined AEBITDA grew 41% quarter over quarter from $7.1m to $10m (please see reconciliation below) despite pro forma, combined revenue dipping 3.5% sequentially.
  • Management confirmed that all of the pre-merger actions promised by the previous StarTek team in terms of ramping higher margin business wins are actually happening and have started to produce a much improved AEBITDA margin and growth profile.
  • Confirmed strong sales momentum as cross-selling opportunities turn into actual bookings: one cited example was landing a major global retailer for the Tegucigalpa, Honduras campus that is affiliated to another client that had been served by Aegis out of one of their campuses in India.
  • Net Debt as at Q3'18 of $157.8m came in significantly lower than the pre-merger Net Debt Target of $184m.

With StarTek trading under 6x my 2019 AEBITDA estimates and, given the revenue and margin expansion about to ensue, the strategic relationship with Amazon (AMZN), the likely re-introduction into the Russell 2000 index in 7 months, and the still stubbornly high short interest balance, I continue to see at least a 50% to 80% increase in the stock price over the next 12 months and a 150% to 200% price increase over the next 18 to 24 months; with a lot less risk.

Q3'18 Revenue, GAAP/Non-GAAP EPS, and Adjusted EBITDA Reconciliation

Given that the transaction closed on July 20, 2018, the reported Q3'18 results exclude the 19-day stub period from July 1 to July 19, making the reported results "messy" and requiring a bit of work to find the pro forma, full quarter numbers within the information provided in the last Form 8k and Form 10Q filed by the company. Please notice the detail in the numbers and the corresponding notes below.

3 months ended September 30, 2018
Reported Full Quarter2
Revenue $151.5m $163.9m
GAAP Net Income (Loss) ($10.9m) ($8.9m)
GAAP EPS ($0.29)1 ($0.24)1
Pro Forma, Full Quarter, Combined Non-GAAP EPS Comparison
Q3'182 Q2'186,7
Revenue $163.9m $169.9m
GAAP Net Income (Loss) ($8.9m) ($7.3m)
Plus Nonrecurring Charges
+ Restructuring Charges $2.6m3 $0.5m
+ Transaction Fees $3.9m3 $1.0m
+ One-time Charges $0.6m4 $0
+ Provision for Doubtful Accounts $1.7m5 $0
Non-GAAP Net Income (Loss) ($0.1)m ($5.8m)
Non-GAAP EPS $0.001 ($0.16)1
Pro Forma, Combined Adjusted EBITDA for Q2 2018
Aegis6 Startek7
Revenue $110.2m $59.7m
Operating Income (Loss) $1.6m ($3.3m)
+ Depreciation & Amortization $4.7m $2.3m
+ Restructuring Charges $0 $0.5m
+ Transaction Fees $0 $1.0m
+ Share-based Compensation $0 $0.2m
Adjusted EBITDA $6.35m $0.8m
Combined Revenue $169.9m
Combined Adjusted EBITDA $7.1m
Combined Adjusted EBITDA Margin 4.2%
Estimated Pro Forma, Combined Adjusted EBITDA Comparison
Q3'182 Q2'186,7
Revenue $163.9m $169.9m
GAAP Operating Income (Loss) ($5.6m)8 ($1.7m)
+ Restructuring Charges $2.6m3 $0.5m
+ Transaction Fees $3.9m3 $1.0m
+ One-time Charges $0.6m4 $0
Non-GAAP Operating Income (Loss) $1.5m ($0.2m)
+ Depreciation & Amortization $8.2m9 $7.0m
+ Share-based Compensation $0.3m $0.2m
Estimated Pro Forma, Combined Adjusted EBITDA $10m10 $7.1m
Estimated Pro Forma, Combined Adjusted EBITDA Margin 6.1% 4.2%
Q3'18 Adjusted EBITDA Potential
Estimated Pro Forma, Combined Adjusted EBITDA $10m $7.1m
+ Provision for Doubtful Accounts (non-cash charge) $1.7m5 $0
Potential Pro Forma, Combined Adjusted EBITDA $11.7m $7.1m
Potential Pro Forma, Combined Adjusted EBITDA Margin 7.1% 4.2%

1) Based on 37.06 million shares outstanding as of 10/31/2018. Source: Form 10Q filed 11/09/18

2) Pro Forma unaudited including stub period. Source: Form 10Q filed 11/09/18, page 11

3) As reported excluding stub period. Pro Forma number for full quarter not provided.

4) SG&A expenses also included a one-time charge of $0.6 million for bonus and severance paid to certain executives. Source: Form 10Q filed 11/09/18, page 22

5) The Statement of Cash Flows shows a $1.726m non-cash Provision for Doubtful Accounts charge related to legacy client accounts served by Aegis (some of these are government clients that tend to take longer to pay than private sector clients). As part of the combination transaction, the company adopted a policy to provision against account receivables older than one year. Management indicated that they remain optimistic that a large part of this provision may be recovered in the near term. Source: Form 10Q filed 11/09/18, page 4

6) As reported on Form 8k filed on 11/08/18 and 10Q filed on 11/09/18. Subtracting 3 months Sept.'18 numbers from 6 months Sept.'18 numbers yield Aegis-only June'18 numbers.

7) As reported on Form 8k filed 08/07/18 for Q2'18 earnings release.

8) Adjusting the reported ($6.6m) of operating loss that excluded the stub period taking into account the $2m difference of additional net income when including the stub period. See note 2 above. Since there is no way to know how much of the $2m impacts operating income and how much would be below the operating income line, I am conservatively applying 50% above the operating income line (adding $1m into the gross profit) and 50% below. Pro Forma number for full quarter not provided.

9) As reported on Form 8k filed on 11/08/18: $7.919m; plus an estimate for the StarTek-only stub period of $0.3m based on the reported Q2'18 D&A amount. Pro Forma number for full quarter not provided.

10) This an estimated number given that the actual, pro forma number was not provided and both the GAAP operating income and D&A numbers for the full quarter are also estimates based on information provided in the 10Q, as explained in notes 8 and 9 above.

Q3'18 Analysis

Full Q3'18 pro forma, combined revenue was $163.9m declining 3.5% from the previous quarter on a USD basis. Despite this revenue dip, estimated, pro forma AEBITDA increased 41% from $7.1m to $10m sequentially; with AEBITDA margins increasing sharply from 4.2% to 6.1%. Management indicated that strong revenue growth from the non-telecom verticals with higher gross margins, as well as improved capacity utilization, contributed to the strong AEBITDA growth during the quarter.

As disclosed on page 11 of the 10Q, revenue included for the StarTek side since the acquisition date was $45.5m; so revenue for the Aegis side was $106m for Q3'18 ($151.5m - $45.5m). This $106m in revenue represents an 11.5% decline from the $119.8m achieved by Aegis during the same quarter last year due to a $13.5m F/X impact from the Argentine Peso and the Indian Rupee depreciation against the USD (page 20,21). The Indian Rupee depreciated 11% and the Argentine Peso depreciated 130% year over year against the USD. Without this foreign currency fluctuation impact, Aegis' Q3'18 revenue would have been $119.5m; essentially flat on a constant currency basis, year over year. It is important to point out that the F/X impact does not result in a loss of margin in local currency as the local operations are mostly naturally hedged; both the revenues and costs are received and paid in local currency. The challenge becomes when converting the depreciated local currency into a stronger USD for accounting and reporting purposes. Moreover, having the majority of the debt denominated in USD also becomes problematic when the cash inflows denominated in a depreciated currency are needed to make cash payment outflows denominated in an appreciated currency. To this end, management confirmed that, when it becomes advantageous to refinance their loan facilities, one of their goals is to have facility tranches available in the different local currencies in which they operate in order to achieve a natural hedge when it comes to debt service.

Also disclosed on page 11 of the 10Q, the pro forma full quarter net loss was ($8.9m) as compared to the net loss reported of ($10.9m) which excluded the 19 day stub period from July 1 to July 19. No further information was provided in terms of the additional $2m in net income as it relates to what income statement lines it belongs to: how much of it above the operating income line (higher gross profit, lower SG&A, or a combination of both?) and how much below the operating income line. So, here, I have to make an educated guess as to where to apply the additional profit from the legacy StarTek-only stub period. I am conservatively applying 50% (or $1m) above the operating income line and 50% under that line. Given that the SG&A line was already pretty heavy, I am applying the entire additional $1m in operating income to the gross profit line. The result is an estimated gross margin of 14.5% marking a sequential improvement over the estimated 13% pro forma, combined gross margin achieved the previous quarter.

SG&A came in at a pretty high $22.8m or 14% of revenue for the quarter, but dissecting this number further will give us a good indication of what to expect going forward. As disclosed on page 22 of the 10Q, this total included a one-time charge of $0.6m for executive severance. In addition, I believe the $1.7m Provision for Doubtful Accounts charge shown in the Statement of Cash Flows (page 4) was also included in the SG&A line (as GAAP would usually call for); this non-cash charge relates to a policy adopted during the combination transaction to account for receivables older than one year. Although reserving against accounts receivables that become late in payment is a normal, recurring part of any business, it is usually accrued for much earlier after 90 or 120 days for the different amounts during that specific period; and not a full amount accumulated for over a year in one big charge. For this reason, I believe it would be prudent to treat it as a special, one-time charge to arrive at a better base SG&A measure; stripping both of these charges gets us to a $20.5m SG&A base declining $1.7m or 7.7% sequentially vs. the prior quarter.

Adjusting for these one-time, nonrecurring charges as these two entities integrate and including the stub period for a pro forma, full quarter view show a much different picture than what had to be reported under GAAP. As shown in the non-GAAP EPS reconciliation table above, non-GAAP net income grew by $5.7m resulting in an essentially breakeven $0.00 cents per share in Q3'18 as compared to a non-GAAP loss of ($0.16) cents per share the previous quarter; all as a direct result from gross margin improvements and lower SG&A base.

Although the estimated AEBITDA improved nicely to $10m sequentially as presented in the reconciliation above, it is important to point out that this number includes the impact of the sizable, non-cash Provision for Doubtful Accounts charge described above. I believe adjusting for this special, non-cash charge would give us a better measure of the potential AEBITDA generation capability embedded within their operating model. Without that charge, the potential estimated AEBITDA would have been at least $11.7m for the quarter with an estimated AEBITDA margin of 7.1%.

Link to my reformatted SRT Pro Forma, Combined P&L and Cash Flow for Q2 and Q3 2018

Forecast

Based on these hidden bright spots, I point out above regarding the underlying business and on some very key comments made by management during the earnings call, I expect continuous AEBITDA improvement quarter over quarter going forward; with measured improvement over the next 2 to 3 quarters leading to more accelerated improvement into the 2nd half of 2019.

Some of the key comments supporting continued AEBITDA improvement as early as the current quarter Q4'18:

  • Margins for Q3 reflect just a small amount of the $12.5m in annualized cost savings implemented, with the majority to be seen in Q4 onwards.
  • Seeing strong growth in the non-telecom verticals; and most of the decline in telecom, particularly on the U.S. side, has already happened through the system.
  • Recently started ramping business that is expected to drive 1000+ seats over the next 4 to 6 months increasing capacity utilization across the globe.

Over the next 12 months, I expect revenue to grow moderately, mostly from the legacy StarTek side, as the higher margin new business won pre-merger continues to ramp improving capacity utilization. I expect the SG&A line to range around $22m per quarter, despite the savings announced, as the company uses some of the savings to invest in sales and business development; however, as revenues grow, the SG&A amount as a percentage of revenue should improve nicely. The improved utilization, coupled with the full impact from the implemented cost savings and existing operating leverage embedded within the organization, will result in higher AEBITDA and AEBITDA margins going forward.

Link to SRT P&L and Cash Flow Forecast

As shown in the attached P&L and cash flow forecast above, I expect the company to be marginally profitable on a non-GAAP basis this December 2018 quarter. And, if they are able to recover any part of the account receivables that were provisioned for in Q3'18, any recovery would go straight to the bottom line for a potentially even higher earnings upside. I expect AEBITDA to also grow sequentially as a direct result from the points mentioned above.

Annual Revenue, AEBITDA, and Net Debt Forecast:

2018 Revenue

NTM Revenue

2019 Revenue 2020 Revenue 2021 Revenue 2022 Revenue
$693m $723m $740m $785m $832m $882m
Annual Growth Rate Growth Rate vs 2018 Annual Growth Rate Annual Growth Rate Annual Growth Rate Annual Growth Rate
-9.9% 3.9% 6.8% 6.1% 6.0% 6.0%
2018 AEBITDA

NTM AEBITDA

2019 AEBITDA 2020 AEBITDA 2021 AEBITDA 2021 AEBITDA
$50m $60m $65m $77m $83m $97m
AEBITDA Margin NTM AEBITDA Margin AEBITDA Margin AEBITDA Margin AEBITDA Margin AEBITDA Margin
7.2% 8.3% 8.8% 9.8% 10.0% 11.0%
2018 CAPEX NTM CAPEX 2019 CAPEX 2020 CAPEX 2021 CAPEX 2022 CAPEX
$14.3 $16.0 $16.0 $16.0 $16.0 $16.0
2018 NetDebt NTM NetDebt 2019 NetDebt 2020 NetDebt 2021 NetDebt 2022 NetCash
($161) ($151) ($142) ($102) ($55) $1.4
Net Debt Leverage NTM Net Debt Leverage Net Debt Leverage Net Debt Leverage Net Debt Leverage Net Debt Leverage
3.2 2.5 2.2 1.3 0.7 0.0

1) CAPEX estimates based on both entities' historical maintenance and expansionary capex. As business becomes more consistent and predictable, I expect management to release annual capex budget.

2) Net Debt estimates based on the company using the available Free Cash Flow after interest expense to pay down debt. Once economies of scale begin to kick in at certain revenue levels, their financial model allows for AEBITDA margins of 10% to 12%. At those levels, the company would be able to use about 70% of AEBITDA generation to accelerate debt reduction.

3) The company continues to reiterate its goal to achieve an incremental $30m in AEBITDA to $80m by 2020 from the $50m in combined AEBITDA achieved in 2017. Achieving annual revenue of $750m to $780m with AEBITDA margins around 10.5% will get them there.

4) I try to be conservative on the forecasts presented and believe the company has enough room to easily outperform the numbers above.

Valuation

As shown below, at around $6.35 per share, the stock is currently trading under 6x my 2019 AEBITDA estimates; and only at 4.4x 2020 estimates. I would remind investors that the company continues to reiterate their goal to achieve an incremental $30m in annual AEBITDA to $80m by 2020 from the $50m in combined AEBITDA achieved in 2017. Hopefully, I have been able to show here that their path to achieve that goal is pretty clear and straightforward. Achieving annual revenue of $750m to $780m with AEBITDA margins around 10.5% would carry them to that goal. In fact, the Amazon strategic partnership alone, with the expectation of that relationship generating anywhere between $75m and $125m in annual revenue, could get them to those revenue levels within a year.

As the company performs and achieves the expected revenue and AEBITDA growth rates and results from the Amazon relationship become evident, StarTek shares, in my view, should again trade closer to 8x to 9x EV/AEBITDA multiples over the next twelve months. At those multiples and based on my conservative AEBITDA estimates, I believe SRT shares could reach $9.00 to $11.00 per share over the next 12 months and potentially reach the $15 to $20 range within the next 24 to 36 months.

Current Stock Price: $6.35
Current Stock Price AEBITDA Multiple
NTM 2019 2020 2021 2022
6.4 5.8 4.4 3.5 2.4
Expected Market AEBITDA Multiple based on growth and margins
NTM 2019 2020 2021 2022
8.5 8.5 8.5 8.5 8.5
Stock Price Target based on Expected AEBITDA Multiple
NTM 2019 2020 2021 2022
$9.67 $11.12 $14.85 $17.59 $22.29
Shares Outstanding: 37.1m

Catalysts

  • Performance and AEBITDA Growth: I expect to see improved results quarter over quarter as early as the current December 2018 quarter to be released sometime in March.
  • Russell 2000 Index Reconstitution: As I have presented on my previous article, market cap simple math gives us a very high probability that StarTek would rejoin the Russell 2000 index once it reconstitutes in June 2019. We should see action as it relates to this probability as early as April when it becomes clear that the market cap surpasses the minimum threshold; the minimum market cap to join the index in June 2018 was $159m. Lists of companies targeted to be added to the index begin to come out in May.
  • Short Interest: The short interest balance remains stubbornly high at 928k shares as of 11/15/18. I estimate that the average cost of that position is somewhere around the $7 per share level. It seems to me that their window to cover profitably is shrinking based on the other catalysts listed above. Once end of year tax loss selling is done and the New Year comes, we could see a race to cover before Q4'18 results are out in March; this potential covering, whenever it happens, may itself become a catalyst due to the relative illiquidity of this stock given its small float.
  • Evidence of Amazon revenue starting to build as warrants start vesting after reaching certain revenue thresholds. Any given quarter going forward, we should see warrants vesting for Amazon as the company reports the warrant accounting in their results. Additionally, in any given quarter going forward, Amazon could be the only client that may reach the 10% of revenue threshold; this would have to be disclosed in the company's 10Q. I would consider this as extremely positive.
  • Inevitable increased analyst coverage given the size: I fully expect the CEO and CFO to schedule more investor conferences and engage with more analysts after results improve to a more predictable and stable level over the next 2 to 3 quarters. This story will not remain under the radar for long.

Conclusion

Looking under the hood of StarTek's Q3'18 results reveals, in my view, tangible evidence that this combination was truly compelling and that the benefits of such are about to start presenting themselves soon. Take advantage of tax loss selling before the coming catalysts presented above start to fire this thing up; I certainly have.

Within the next 3 years, I believe StarTek will become a very sought after premium asset due to its truly diversified nature, truly global reach, and a unique 8 year strategic partnership with such a powerful client as is Amazon. There are many larger companies in the space that do not have the global presence, especially in Asia, that StarTek now has. In fact, the only company I see that has a truly global delivery platform is Teleperformance (OTCPK:TLPFY, OTCPK:TLPFF), a behemoth in the BPO industry space with a market cap of over $9 billion. Once StarTek is fully integrated, and AEBITDA margins in the 10% to 12% range are the norm, while boasting a client service agreement with Amazon with 4 to 5 years left in it, larger players will see a great opportunity to add an invaluable and complementary Asia piece to their platform with many large, marquee clients to boot; and all of this with very accretive earnings. Only if Teleperformance allows that to happen, that is. This would obviously be an event for a few years down the line still, but a transaction for the gem that StarTek is positioned to become will easily command a 10x EV/AEBITDA multiple; and as they say, the early bird gets the worm.

DISCLAIMER: Although the information contained herein has been obtained from sources that Omar A. Samalot believes to be reliable, he does not guarantee its accuracy and such information may be incomplete or condensed. All opinions and estimates included herein constitute his judgment as of this date and are subject to change without notice. This report is for informational purposes only and is not intended as an offer or solicitation with respect to the purchase or sale of any security. He, or persons involved in the preparation of this report, may from time to time have long or short positions in these securities, and may buy or sell these securities.

Copyright © Omar A. Samalot, 2018

Disclosure: I am/we are long SRT.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: Omar A. Samalot is an independent financial analyst, owns SRT shares as of the date of this report, and has increased his position since the Q3'18 earnings report based on news, announcements, and his own analysis. The content of this report cannot be copied or distributed without the author’s written consent.

Editor's Note: This article discusses one or more securities that do not trade on a major U.S. exchange. Please be aware of the risks associated with these stocks.