Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

Hollysys Automation: Why I Sold My Shares

|Includes: HollySys Automation Technologies, Ltd. (HOLI)

I posted my first Instablog comment/analysis on Hollysys Automation ("HOLI") on 13 December 2013. In it I explained why I believed a detailed analysis of the company's accounts was necessary to understand the true performance of the company. I explained that investors should pay particular attention to how the company reports earnings (on a "non-GAAP" basis), on the level of "VAT rebates and government subsidies" it gets in China, on the impact of recent acquisitions and, finally, on the appreciation of the RMB vs. the US$.

Based on these analysis criteria, on the recently published 2Q14 results (http://ir.hollysys.com/2014-02-17-Hollysys-Automation-Technologies-Reports-Unaudited-Financial-Results-for-Fiscal-Year-2014-Second-Quarter-Ended-on-December-31-2013) and the recent strength of the share price, I have decided to sell my shares (netting a small profit). I now firmly believe that the stock is overvalued and that analysts keep pushing the stock based on a poor understanding of the business or because they are not paying sufficient attention to the underlying fundamentals.

1/ 2Q14 results analysis

Hollysys reported 2Q14 results (ending 31 December 2013) on 17 February 2014. The "Q2 Financial Highlights" from the press release make for impressive reading, highlighting a 90.7% year-on-year increase in Non-GAAP net income to US$25.9m, and a year-on-year 40% increase in backlog to US$503.3m. On the face of it the company appears to be undergoing extremely strong growth. But let's take a closer look.

- Net income attributable to Hollysys shareholders for 2Q14 was actually US$18.3m, showing growth of 40.1% year-on-year. This is a big difference with the 90.7% year-on-year growth highlighted in the press release. But what is even more troubling to me is that it is compared to 2Q13 net income which was an extremely weak basis of comparison (at US$13.071m it was actually down 35.5% year-on-year vs. 2Q12). If we go back 2 years and look at 2Q12 net income you'll note that it stood at US$20.271m. So 2Q14 net income has only partially made up for last year's shortfall, but not fully.

- Moreover, 2Q14 net income now includes a contribution from the BOND group, which was acquired and consolidated from 4Q13 last year. Therefore an adjustment needs to be made to reflect this if one wants to look at net income growth on a like-for-like basis. I estimate that BOND may have contributed c.US$1.5-2m in net income in 2Q14, but this is just a guess...the company didn't disclose this information.

- In addition, 2Q14 net income (and all of Hollysys's accounts in fact) should be converted back to RMB to adjust for currency fluctuations (the company reports in US$, while it operates mostly in RMB). When the company reported net income of US$20.271m in 2Q12 for example the RMB/US$ exchange rate was around 6.23 (implying 2Q12 net income in RMB was roughly RMB126.3m). For 2Q14 the exchange rate was roughly 6.055 (implying 2Q14 net income in RMB was roughly RMB110.8m, 12.2% lower than 2Q12).

- Finally, and as I already highlighted in my December 2013 Instablog, Hollysys's profits seem to be increasingly reliant on growing "VAT rebates and government subsidies". These stood at US$10.415m in 2Q14, up 42.8% year-on-year, and represented 40% of Hollysys's "Income from operations". Note however that "VAT rebates and government subsidies" amounted to US$13.456m in 2Q12 (the highest ever for the company!).

- The backlog analysis should, first and foremost in my view, highlight the company's organic growth. So when Hollysys states that backlog has grown 40% year-on-year it makes it sound as though the company has received plenty of new orders, operates on a fast growing market, and that it has bright growth prospects.

- The reality, however, is that the US$106.5m in M&E backlog comes from consolidating the BOND group within the company's accounts and in no way reflects Hollysys's traditional business in China (BOND operates in Malaysia and Singapore, and in another field called 'Mechanical and Electrical solutions'). So, when we strip out BOND's contribution to the group backlog, Hollysys's backlog at the end of 2Q14 (comparable to last year's scope of business) stands at US$396.8m, up 10.3% year-on-year.

- In addition, if we adjust the US$396.8m backlog figure for the RMB/US$ exchange rate, then year-on-year organic growth stands at 7.1%...which is much lower than the 40% reported and gives a clearer picture of the company's new order growth in the Chinese market.

2/ Profitability analysis

Hollysys's profitability level has recovered a bit from last year's trough, but still remains below that of the previous years. This is because the company has to increasingly invest to maintain margins and find growth, especially through acquisitions, and so its marginal returns on new investments are decreasing. This is normal and is the case with nearly all businesses.

In the case of Hollysys we can point out the following:

- The company's Equity stood at US$314.435m in 2Q12, and has since grown to US$459.013, +46% over the past 2 years. Likewise the company's Total Assets stood at US$509.414m in 2Q12, and have since grown to US$865.791, +70% over the past 2 years.

- Over the same time period (2Q12 to 2Q14), Income from operations grew 29%, and Net income grew 15.6%. Return on Equity, a measure of profitability to investors, has therefore fallen noticeably - from 21% to 16.7%. As for the Return on Assets, a measure of profitability of the overall business, it has fallen from 13% to 8.9%.

Although RoE and RoA have been falling over the past 2 years, they remain at healthy levels. The reason I highlight these numbers is for 2 reasons: 1/ to show a trend and underline how quickly profitability is decreasing and 2/ to give you, the reader, an idea of low profitability could be if Hollysys could no longer rely on as much "VAT rebates and government subsidies", which represented c.40% of its Income from Operations in the past 2 years.

3/ Valuation comment

Valuation is always a tricky and subjective issue. And most investors like easy approaches like looking at comparable companies. So, in the case of Hollysys, I often hear how cheap it is when we compare its P/E ratio with the likes of Emerson Electric Co. ("EMR"), a multinational company with a US$46bn market cap…hardly comparable if you ask me…but that's the way it goes.

As for me, I just like to look at simple ratios which usually tell me how much of a premium or a discount I am paying relative to how much has been invested in the company already. I find it makes more sense and gives you an indication of how bullish or bearish the market actually is.

- Hollysys shares are currently trading around US$20.00, implying a market cap of US$1.15bn. Given the Net cash sitting on the balance sheet at 2Q14 this implies an Enterprise Value of US$1.045bn.

- That means that the Price/Book ratio is currently equal to 2.5x implying that the market is already discounting that the company will make at least a cumulative future net income of US$691m…which is quite a lot if you ask me, and leaves little room for disappointment.

- It also means the Enterprise Value is equal to 2.71x the Capital Employed (Fixed assets + working capital). This is a high ratio because it means that, if you were to buy this company at this price, you would effectively be paying an amount equal to 2.71x what has been invested in it to make it run.

I have seen stocks trade on much higher valuation ratios in the past so I am not shocked by Hollysys's current multiples. However, I certainly don't find them to be good value at this level, especially when considering the risks involved. This is why I just sold my shares.

3/ What to expect short to medium term

Hollysys has recently increased its guidance for the year ending June 2014, which partly explains the stock's recent performance. If you pay close attention, however, you will note that guidance is for "non-GAAP net income"…which means it is difficult for them to miss since "non-GAAP net income" can include or exclude pretty much whatever they want. That's why it's called "non-GAAP".

In the short term I would expect a good set of results published for 3Q14. Again this is because 3Q13 offered a weak basis of comparison, because the RMB (which traded at 6.2241 on average in 3Q13) has further appreciated (trading at an average 6.0634 so far in 3Q14, i.e. +4%), and because the BOND acquisition was not consolidated yet in 3Q13. "non-GAAP net income" growth should therefore look very good in 3Q14.

4Q14 might look less strong though. Indeed, the BOND group was already consolidated in 4Q13 so its consolidation in 4Q14 won't add very much, maybe even less than last year if it's not doing as well, and if the currencies it operates in (Malaysian Ringgit and Singaporean Dollar) keep falling. In addition, the RMB has started to depreciate versus the US Dollar this past week. If this trend continues Hollysys will no longer enjoy the conversion gains it previously had in its accounts, which means it will get harder to show growth in earnings, or backlog, in US Dollars.

But 4Q14 accounts are still a long way away. And by then nobody will remember reading this Instablog. It is also possible that Hollysys makes more acquisitions in the coming months, further adding to positive sentiment in the market.

Finally, close attention should be paid to backlog growth in the coming quarters to understand the company's prospects for 2015 (Bloomberg consensus points to 20% EPS growth in 2015). Indeed, 2Q14 backlog is already down 2.4% vs. 1Q14 backlog and down 6.2% if we exclude BOND's business. And I suspect this could decrease further if the Chinese government were to become serious about restraining credit in the economy, which would adversely impact corporate demand for industrial automation. Moreover, I doubt large government budgets to build high speed railways will go on forever either.

Whatever the case, I have sold my shares…and made it (I hope!) pretty clear why this was the case.