SpartanNash: LIFO Adjustments, A Step Too Far

Summary

• SpartanNash offers an attractive 4.11% dividend yield but has declining EPS estimates limiting potential for increase in P/E ratio, and thus limiting share price growth.
• Some of the adjustments made in arriving at adjusted non-GAAP results appear difficult to justify, leading to a question mark over the level of reported non-GAAP EPS.
• A wait and see approach is likely appropriate, to ascertain if management's optimism about future growth can be realized.

Investment Thesis

In my article published September 29, 2021, I wrote,

At current share price SpartanNash Company (NASDAQ:SPTN) offers an attractive 3.79% dividend yield. The current P/E ratio of 9.29 is below its historical levels of around 10.0 to 10.5. Unfortunately, SA analysts' EPS estimates indicate a decline in EPS through end of 2023, so any change in P/E ratio is more likely to be a decrease than an increase. With a prospect for declining EPS and declining P/E ratio causing further share price declines, total returns through end of 2023 could be negative, despite the dividend yield. I believe there will likely be opportunities in the fourth quarter for lower stock prices across the board, including SpartanNash, as I believe a market correction is overdue. At a lower share price the dividend yield should be higher and the risk of share price decline lower, at which time SpartanNash might come into buy range.

At market close on February 23, 2024, the SpartanNash share price was down 5.85% on the price at date of the abovementioned article, and the dividend yield had increased from 3.79% to 4.11%. Per Seeking Alpha Premium, total return over this period was 1.37% due to the dividend receipts exceeding the fall in the share price.

I set out to find if the lower share price and the higher dividend yield would now justify an upgrade to a Buy rating. Unfortunately, Table A below reveals some discouraging trends.

Table A

Comments on Table A -

• GAAP earnings of $52 million for 2023 are below 2016 level of $57 million.
• GAAP earnings in the intervening years average ~$29 million per year.
• Adjusted non-GAAP earnings of $76 million for 2023 are below 2016 level of $82 million.
• Adjusted non-GAAP earnings in the intervening years average ~$73 million per year, below 2016 level.
• Expenditures of $182 million over the 8-year period on restructuring, transformation, and reorganization should have resulted in improvements in Adjusted non-GAAP earnings, but this is not the case.
• Regularly recurring Restructuring, transformation, and reorganization expenses appear to be a normal part of operations, which raises the question whether they can legitimately be added back to normalize the result.
• The add back of LIFO expense in the calculation of Adjusted non-GAAP earnings is entirely inappropriate. This is an attempt to include in adjusted earnings the one off gains on the value of inventories due to price inflation. LIFO accounting ensures the current price of goods sold is reflected in cost of goods sold. Therefore, GAAP results based on LIFO accounting already reflect the "true" underlying cost of goods sold and no adjustment is required to normalize the result for non-GAAP purposes.
• Adjusted non-GAAP EPS of $2.18 for 2023 has the benefit of a reduction of 2.7 million in share count since 2016. Without this the result would be an EPS of $2.02, compared to $2.19 for 2016. The reduction in share count is not increasing EPS, merely preventing the EPS from decreasing further.
• The reduction in share count has not been achieved by using earnings in excess of dividends to repurchase shares. Balance sheet analysis further below shows funding of repurchases of shares is totally attributable to increased borrowings.
• Transparency has been reduced by the decision in 2022 to no longer separately report segment operating earnings (losses) for Military, at a time when Military was showing increasing operating losses year by year.

Based on the foregoing review, and my detailed analysis below, I am inclined to downgrade SpartanNash from Hold to Sell. On the other hand, if the management enthusiasm displayed in the Q4-2023 earnings call and presentation can be translated into real growth in sales and earnings, that would likely result in an increase in the P/E ratio which together with higher EPS could drive good share price gains above the current depressed share price. Add to that the over 4% dividend yield, and a continuing Hold recommendation, with a wait and see approach is likely the most appropriate course of action.

My usual detailed structured financial analysis follows below.

Looking For Share Market Mispricing Of Stocks

What I'm primarily looking for here are instances of market mispricing of stocks due to distortions to many of the usual statistics used for screening stocks for buy/hold/sell decisions.

I believe the answer is to compare projections, based on analysts' estimates out to the end of 2024 or later, to past performance. Summarized in Tables 1 and 2 below are the results of compiling and analyzing the data on this basis.

Table 1 - Detailed Financial History And Projections

Comments on Table 1 -

Performance over past 7 years - Table 1 shows SpartanNash non-GAAP EPS has barely changed over the past seven years, from $2.19 for 2016 to $2.18 for 2023. Over the same period, SpartanNash's stock price declined by an average (7.48)% per year, from $39.54 at the end of 2016 to $22.95 at the end of 2023. This is primarily due to a decrease in the P/E multiple from 18.05 at the end of 2016 to 10.53 at the end of 2023. Despite the lack of EPS growth, dividend per share has grown by an average yearly rate of 5.28% from $0.60 at the end of 2016 to $0.86 at the end of 2023. This dividend growth rate of 5.28% per year, despite the absence of EPS growth, has resulted in an increase in the dividend payout ratio from 27.4% for 2016 to a still modest 39.4% at the end of 2023. The decline in the share price over the past seven years has been the primary reason for the dividend yield increasing by around 2.5 times, from 1.52% at the end of 2016 to 3.75% at end of 2023. Further decline in the share price through February 23, 2024 has resulted in the dividend yield further increasing to 4.11%.

Modeling for targeted rates of return - Table 1 documents historical data from 2016 to 2019, including share prices, P/E ratios, EPS and DPS, and EPS and DPS growth rates. The table also includes estimates out to 2026 for share prices, P/E ratios, EPS and DPS, and EPS and DPS growth rates (note - while estimates are shown for analysts' EPS estimates out to 2024, 2025 and 2026 where available, estimates do tend to become less reliable, the further out the estimates go. These estimates are only considered sufficiently reliable if there are at least three analysts' contributing estimates for the year in question). Table 1 allows modeling for target total rates of return. In the case shown above, the target set for total rate of return is 7.5% per year through the end of 2025 (see line 12), based on buying at the February 23, 2024 closing share price level. As noted above, estimates become less reliable in the later years. I have decided to input a target return based on 2025 year, which has EPS estimates from three analysts. The table shows to achieve the 7.5% return, the required average yearly share price growth rate from February 23, 2024 through December 31, 2025, is 3.15% (line 53). Dividends and dividend growth account for the balance of the target 7.5% total return.

SpartanNash's Projected Returns Based On Selected Historical P/E Ratios Through End Of 2025

Table 2 below provides additional scenarios projecting potential returns based on select historical P/E ratios and analysts' consensus, low, and high EPS estimates per Seeking Alpha Premium through end of 2025.

Table 2 - Summary of relevant projections SpartanNash

Table 2 provides comparative data for buying at closing share price on February 23, 2024 and holding through the end of year 2024 to 2026 (where estimates are available). There is a total of twelve valuation scenarios for each year, comprised of three EPS estimates (SA Premium analysts' consensus, low and high) across four different P/E ratio estimates, based on historical data, plus a fourth P/E ratio selected to provide an alternative scenario. SpartanNash's P/E ratio is presently 9.60, which is well below the historical average P/E ratio of 11.12. Table 2 shows potential returns from an investment in shares of the company across the range of P/E ratios This analysis, from hereon, assumes an investor buying SpartanNash shares today would be prepared to hold through 2025, if necessary, to achieve their return objectives. Comments on contents of Table 2, for the period to 2025 column follow.

Consensus, low and high EPS estimates

All EPS estimates are based on analysts' consensus, low and high estimates per SA Premium. This is designed to provide a range of valuation estimates ranging from low to most likely, to high based on analysts' assessments. I could generate my own estimates, but these would likely fall within the same range and would not add to the value of the exercise. This is particularly so in respect of well-established businesses such as SpartanNash. I believe the "low" estimates should be considered important. It's prudent to manage risk by knowing the potential worst-case scenarios from whatever cause.

Alternative P/E ratios utilized in scenarios

1. The actual P/E ratios at the share buy date are based on actual non-GAAP EPS for FY-2023.
2. A modified average P/E ratio based on 30 quarter-end P/E ratios from Q4 2016 to Q4 2023 and current P/E ratio in Q1 2024. The average of these P/E ratios has been modified to exclude the three highest and three lowest to remove outliers that might otherwise distort the result.
3. A modified low P/E ratio was calculated using the same data set used for calculating the modified average P/E ratio, and calculated on a similar basis, with the three highest and lowest P/E ratios excluded.
4. A median P/E ratio is calculated using the same data set used for calculating the modified average P/E ratio. Of course, the median is the same whether or not the three highest and lowest P/E ratios are excluded. In the case of SpartanNash, I have chosen to use an assumed P/E ratio of 18.0 in place of the historical median of 11.01 (similar to the average). I have done this to provide an idea of the impact on returns of the multiple increasingly significantly above the present level and the historical average. The selected P/E multiple of 18.0 compares to the sector median of 18.02 for PE Non-GAAP [FWD], per Seeking Alpha Premium metrics.

Reliability of EPS estimates (line 18)

Line 18 shows the range between high and low EPS estimates. The wider the range, the greater disagreement there is between the most optimistic and the most pessimistic analysts, which tends to suggest greater uncertainty in the estimates. There are 3 analysts covering SpartanNash through end of 2025. In my experience, a range of 9.6 percentage points difference in EPS growth estimates among analysts is high, suggesting a considerable degree of uncertainty, and thus increased unreliability.

Projected Returns (lines 18 to 39)

Lines 25, 35 and 45 show, at a range of historical P/E ratio levels, SpartanNash is conservatively indicated to return between negative (4.8)% and positive 4.6% average per year through the end of 2025. The negative (4.8)% return is based on analysts' low EPS estimates and the 4.6% on their high EPS estimates, with a 0.2% return based on consensus estimates. Those are the lowest of the returns under the consensus, low and high EPS scenarios and are based on the P/E ratio remaining at the current level. At the high end of the projected returns for SpartanNash, the indicative returns range from 31.4% to 44.6%, with consensus 38.3%. These returns would require the P/E ratio to increase from the present 9.60 to the ~18.05 at end of 2016. Coincidentally, SpartanNash's sector median P/E non-GAAP [FWD] ratio is 18.02 per SA Premium. However, sector median percentage growth rates for Revenue, EBITDA, and EPS are in the mid to high single digits. Unless SpartanNash can show growth in these areas, an 18.0 P/E ratio looks like an impossible dream. If the P/E ratio were to increase to the historical average of 11.12 by end of 2025, indicative returns would be in the range of 4.3% to 14.6%, with consensus 9.7%. The difference between best and worst cases is an indication of the degree of certainty in analysts' estimates.

Checking SpartanNash' "Equity Bucket"

Table 3.1 SpartanNash Balance Sheet - Summary Format

Over the seven years from year end 2016 to the end of 2023, SpartanNash's shareholders' equity has decreased by $47 million. This reduction is reflected as an increase in debt net of cash of $173 million offset by an increase of $126 million in net assets used in operations. Outstanding shares decreased by 2.9 million from 37.5 million to 34.6 million, over the period, due to share repurchases partially offset by shares issued for stock compensation. Net debt as a percentage of net debt plus equity increased from 33.0% at the end of 2016 to 42.7% at the end of 2023. The $47 million decrease in shareholders' equity over the last seven years is analyzed in Table 3.2 below.

Table 3.2 SpartanNash Balance Sheet - Equity Section

I often find companies report earnings that should flow into and increase shareholders' equity. But often the increase in shareholders' equity does not materialize. Also, there can be distributions out of equity that do not benefit shareholders. Hence, the term "leaky equity bucket." This is happening with SpartanNash.

Explanatory comments on Table 3.2 for the period end FY-2016 to end FY-2023.

• Reported net income (non-GAAP) over the 7-year period totals to $486 million, equivalent to diluted net income per share of $13.47.
• Over the 7-year period, the non-GAAP net income excludes $264 million (EPS effect $7.18) of items regarded as unusual or of a non-recurring nature in order to better show the underlying profitability of SpartanNash.
• Other comprehensive income includes such things as foreign exchange translation adjustments in respect to buildings, plant, and other facilities located overseas and changes in valuation of assets in the pension fund - these are not passed through net income as they fluctuate without affecting operations and can easily reverse in a following period. Nevertheless, they do impact on the value of shareholders' equity at any point in time. For SpartanNash, these items totaled $14 million loss (EPS effect $0.38 loss).
• Amount taken up in equity in relation to shares issued to staff over the 7-years is $47 million. This compares to an estimated market value of $52 million for these shares, resulting in an overstatement of net income by $5 million, which is not material in the context of GAAP net income of $223 million.
• By the time we take the above mentioned items into account, we find, over the 7-year period, the reported non-GAAP EPS of $13.47 ($486 million) has decreased to $5.79 ($205 million), added to funds from operations available for distribution to shareholders.
• Dividends of $194 million, and share repurchases of $121 million were greater than the $205 million from operations, resulting in a shortfall in net income from operations after distributions of $110 million.
• The shortfall of $110 million from operations was offset by the $52 million increase in equity due issues of shares to staff, and $12 million from rights issues, resulting in the decrease of $47 million in shareholders' funds per Table 3.1 above.

SpartanNash: Summary and Conclusions

Table 1 further above shows, in order for SpartanNash to provide total returns of 7.5% through end of 2025, not only must analysts' consensus EPS estimates be met, but the P/E ratio must increase from the current 9.60 to 10.66. A P/E ratio of 10.66 is below historical average P/E multiple levels of around 11.12 (see Table 2). However, given SA analysts' consensus EPS estimates are for EPS to further decline through end of 2025, an increase in P/E ratio appears unlikely. The major issue for SpartanNash is a lack of growth, and no certainty of future growth. At the same time, the over 4% dividend yield and management's confidence in achieving future growth, indicate a Hold with a wait and see approach is likely appropriate.