Understanding Micro-Nano Cap Stocks and Finding Potentially Good Ones

 Small stocks attract many fans, starting with academicians who talk of the “small-cap effect” to individuals who associate small cap with superior growth potential to gunslingers who crave penny stocks. There is genuine merit to going small, but if pursued naively, it can lead to pain. We present here a way to screen for small stocks that doesn’t simply add a small-size rule to the same financial criteria we’d apply to larger stocks.The model, built on  Portfolio123, calls our attention to $EVC, $FNJN, $SORL, $DHX, $CVGI, $NWY, $PTN, $JILL, $ARTX, $AXTI,$ AUDC, $NOA, $ORN, $PRTS, $PFMT, $ASYS, $DX, $SND, $RRD, $NVFY

Step One – What’s the deal with the so-called “Small Cap Effect?”

Eugene Fama and Kenneth French, the famous and widely-imitated factor-investing duo, identified a “small cap effect” (which they called SMB for small minus big), a tendency of smaller stocks to do better. They support their view of the size factor, as they refer to it, the same way they support the other factors they’ve identified; by examining a lot of history and counting a lot of beans.

I’m not into identifying factors based on statistical study, because such conclusions are necessarily limited to the specific characteristics of the tine period(s) studied. I prefer to think in terms of the inherent company characteristics that cause empirical data to come out the way it does. I believe my approach provides a better basis for designing an investing strategy for the unknowable future.

For the most part, small capitalization is associated with small company size (the main exception being when a large company has deteriorated to the point where its stock price sank far enough to cause the issue to look like a small cap). So let’s think in terms of business characteristics that tend to be more prevalent among small fry. To do this, think of two traits: 

• Scale
• Diversification

Large companies typically have both. Small companies tend to have neither.

Scale

Scale, in this context, refers to he role fixed costs play in a company’s income profile. As a simple example, consider a retailer. Cost of merchandise is a variable cost. It moves more or less in line with revenues. In boom times, companies boost their spending for merchandise and wind up selling more goods and getting more revenues. Consider, on the other hand, salaries of headquarters executives and staff. That does not vary in line with sales. Ditto the cost of store space.

More often than not, fixed costs will eat up a larger portion of a small firm’s sales, a diseconomy of scale (in contrast to the more revered idea of economies of scale). We can see the influence of this in the data. 

Gross expenses are not completely variable, but we can think of them that way since most such expenses do vary with revenue. Operating expenses, on the other hand are more fixed than variable. Table 1 shows how this plays out.

Table 1 – Demonstration of Fundamentals for Large and Small Firms

	Median % for companies in . . .		Russell 2000 item as % of Russell 1000 item
	Russell 1000 (bigger)	Russell 2000 (smaller)
Gross Margin TTM	42.33	39.55	93.4%
Gross Margin 5Y Avg.	40.30	38.12	94.6%
Operating Mar. TTM	15.50	8.25	53.2%
Operating Mar 5Y Avg.	12.55	8.66	69.0%
Return on Assets TTM	4.70	1.17	24.9%
Return on Assets 5Y Avg.	4.28	1.33	31.1%
Return on Equity TTM	13.38	6.87	51.3%
Return on Equity 5Y Avg.	12.55	7.33	58.4%

Small companies have moderately lower gross margins, but the falloff in the fixed-cost influenced operating margin is much deeper. And that carries all the way down to returns on assets and equity. 

Tables 2 and 3 demonstrate how changes in sales produce changes in earnings that are much more dramatic for companies more burdened by fixed costs.

Table 2 – Hypothetical Large Company

	Year 1	Year 2	Year 3
Sales	1000	800	1200
% change		-20.0%	50.0%
Gross Profit	403	322.4	483.6
Gross Margin %	40.3	40.3	40.3
Operating Exp	277.5	277.5	277.5
Operating Profit	125.5	44.9	206.1
Oper Margin %	12.55	5.61%	17.18%
Pretax Income	125.5	44.9	206.1
Taxes (21% rate)	26.4	9.4	43.3
Net Income	99.1	35.5	162.8
% change		-64.2%	359.0%

Table 3 Hypothetical Small Company

	Year 1	Year 2	Year 3
Sales	100	80	120
% change		-20.0%	50.0%
Gross Profit	38.1	30.5	45.7
Gross Margin %	38.12	38.12	38.12
Operating Exp	29.46	29.46	29.46
Operating Profit	8.66	1.04	16.28
Oper Margin %	8.66	1.30%	13.57%
Pretax Income	8.66	1.04	16.28
Taxes (21% rate)	1.82	0.22	3.42
Net Income	6.84	0.82	12.86
% change		-88.0%	1471.8%

The impact of fixed costs on the downside is why we say smaller companies are risker. On the other hand, the exaggerated earnings gains fixed costs spark when things turn up, even expectations of this, are why so many investors love small caps. This pair of tendencies, plus the good fortune of history having delivered more good periods than bad, is why the quants, after editing out 20-20 hindsight and calling it statistical significance, say there’s a small-cap effect — but at least they tend to caution that small stocks are apt to be more volatile. 

Diversification

Small companies are also prone to have more volatile sales trends because they typically have less business diversification than larger firms. (Even within supposedly single-industry companies, close readings of 10-K Business Descriptions will show that larger firms can be quite varied in the kinds of goods and services offered and in the nature of their customer bases.) Lesser diversification in business profiles tends to produce more sales volatility, analogous to the way poorly-diversified stock portfolios tend to be more volatile.

Table 4 is a modification of Table 3 and when compared with the latter shows how an increase in sales volatility due to lesser business-portfolio diversification, can really send earnings on some wild rides. 

Table 4 – Hypothetical Small Company, Alternative Sales Trend

	Year 1	Year 2	Year 3
Sales	100	50	135
% change		-50.0%	170.0%
Gross Profit	38.1	19.1	51.5
Gross Margin %	38.12	38.12	38.12
Operating Exp	29.46	29.46	29.46
Operating Profit	8.66	-10.40	22.00
Oper Margin %	8.66	-20.80%	16.30%
Pretax Income	8.66	-10.40	22.00
Taxes (21% rate)	1.82	-2.18	4.62
Net Income	6.84	-8.22	17.38

Here, the impact of the sales change on Year 2 Earnings is so severe, we can’t even compute meaningful percentage changes surrounding it (because we’re going from positive earnings to losses and then back to positive again).

The Bottom Line

When a quant talks about the small cap effect, nod your head “yes,” politely turn away, and roll your eyes. Their reasons are nonsensical, but unless we’re going through or anticipating the imminence of bad times, things are likely to work out such as to allow them to congratulate themselves for studies well done. 

But when you develop a small cap strategy, do so on the basis, not just of size, but of the fundamental characteristics inherent in small size.

I’m not just talking about theoretical purity for the sake of theoretical purity. I’m also talking about actionable results or lack thereof. Although documentation of the small cap effect has been good enough to get studies published in peer-review tenure-track journals, Table 5 shows that it hasn’t been nearly sufficient to allow real-world investors to make enough money to consider the endeavor worthwhile.

Table 5:  Using ETFs To Test Naive Small-Cap Strategies

	Russell 1000 (Small-cap) ETF	Micro-Cap ETF	Russell 1000 (Large-cap) ETF
10-Year Test	IWM	IWC	IWB
Annual % Return	10.21%	9.65%	9.77%
Standard Dev.	20.40%	21.67%	15.53%
Max Drawdown	-55.56%	-55.81%	-49.39%
Beta	1.21	1.24
Annual Alpha %	-0.99%	-1.28%
Avg. 4 wk periods
All	0.65%	0.65%	0.81%
Up markets	3.46%	3.45%	3.14%
Down Markets	-4.77%	-4.75%	-3.68%

Obviously, small alone won’t cut it. The small-cap return is a bit better than the large-cap return, but not enough so to justify the increased risk, and is especially poor in down markets. The micro-cap performance on the start-to-finish 10 year test is a bit worse, and when we get to the rolling test, at this level of decimal rounding, the numbers are the same as for small cap.

So we can’t really buy small just for the sake of buying small. We have to put in some effort to try to tilt the probabilities in favor of small stocks we believe will benefit from the good part of being small, the upside portion of the ihnernetly higher volatility.

Step Two – Stretching Small Into a Full-Blown Strategy

My goal here is to aim very small, at stocks that can be classified as micro caps and smaller. Subject to liquidity tests to be described below. I’m aiming at issues with market caps no higher than $500 million and single-digit stock prices. I’m not serving asset allocators here. I’m aiming at those who really love this stuff — and by the way, if you love investing, these tiny companies are wonderful because you can actually wrap your arms around them without getting overwhelmed by media noise and too many things to look at.

Table 6 sets the context for out work by showing how a naive portfolio of these micro-nano stocks performed.

Table 6:  Testing A Naive Micro-Nano Stock 

	Naive Micro-Nano Cap Strategy	Micro-Cap ETF as benchmark
10-Year Test		IWC
Annual % Return	1.16%	9.77%
Standard Dev.	28.95%	15.53%
Max Drawdown	-64.17%	-49.39%
Beta	1.22
Annual Alpha %	-8.57%
Avg. 4 wk periods
All	0.25%	0.81%
Up markets	4.86%	3.14%
Down Markets	-6.73%	-3.68%

That’s really awful, much worse than the ETFs we looked at, which were. More mundane than bad. We definitely have our work cut out for us. 

The key to screening for small stocks is to not get crazy with guru-like ideas about finding great companies. Small companies often aren’t all that “good,” at least not in terms of fundamentals compared to their larger  brethren. Refer back to Table 1.

So unless you want to reduce your result set to zero or see it loaded with a bunch of aberrant oddballs (oddballs can be nice but aberrant may translate to future outcomes that are less favorable than past performance), you’ll need to relax a bit. Don’t be as picky with your rules as you might be for a larger universe. It’s good to make money on shares of great companies, but never underestimate the investment potential owning shares while the company’s fundamentals improve from dreadful to mediocre, or even better from mediocre to almost palatable. Adopt the aura of ripped jeans and dirty t-shirts, as opposed to the $1,500 suit world of large cap screening. 

I’m still going to use a VQS (Value-Quality-Sentiment) approach based on the ideas set forth in my strategy design course and simple cheat sheet, both of which are available on p123spotlight.com. But I’ll be low-maintenance in how I approach VQS.

Here’s the screen:

• I start by using Portfolio123’s “NOOTC” Universe , which literally means “no OTC” (i.e. eliminate stocks that trade via the pink sheets) in order to steer clear stocks likely to be illiquid and hard, or impossible, to trade
• I then support the liquidity-tradability theme a screening rule that limits consideration to stocks with 60-day average dollars traded (volume times price) of at least $150,00. Don’t kid yourself; this is way sown there in terms of liquidity threshold. But for individuals, it should be ,doable.
• I eliminate ADRs and MLPs. Screening for small stocks is challenging enough. I don’t want extra complexity.
• Next, I define small as
  • • market capitalization no higher than $500 million, and
    • stock price below $10 
• Then, I establish rules for Value an Quality and in the name of ripped jeans-dirty T-shirt flavor, my thresholds are more aimed at eliminating garbage than going for excellence
  • • The stock must score above 50 (out of 100) in the Portfolio123 Basic: Value ranking system (Click here to see details of all Portfolio123 Style-based ranking systems)
    • The stock must score above 25 (out of 100) in the Portfolio123 Basic: Quality ranking system
• Instead of Sentiment, I’ll work here with historic growth, It’s not my favorite thing to use (because past growth doesn’t necessarily  persist into the future). I usually prefer Sentiment as a proxy for the growth factor mandated by VQStheory. But this screen is likely to touch many companies with little or no analyst coverage. So I’ll take my chances on the persistence of growth. That’s one reason why this screen is offered as an idea generator, in which ideas will be studied further, rather than a complete buy-everything model. So here’s the Growth rule
  • • The stock must score above 65 (out of 100) in the Portfolio123 Basic: Growth ranking system (this rank threshold is higher than the others because I figure folks who are interested in companies like these focus heavily on growth)
• Finally, stocks that pass are sorted on the basis of the Portfolio123 QVS (Quality-Value-Sentiment) ranking system and the top 20 are selected for the final list. In this context QVS is very useful.

Use of a multi-style ranking system such as VQS to make my final choice is a technique I really like. It’s a non-judgmental generalist thing. Rather than demand excellence in one thing (like a top 20 position in Value), I’ll let chill and take excellence anywhere I can find it. One stock’s position may be boosted by great Quality metrics, another by super-duper valuation, another by strong Sentiment, and others by tow or three quite OK style scores but one or none that are outstanding, I take this approach because i’m trying to find stocks likely to do well, as opposed to looking for brownie points by proving my value model is better than yours.

Step 3 – Testing the Strategy

While the theoretical underpinnings of what was done are sound (we don’t test for this, we rely on the financial logic explained in the strategy design cheatsheet), we cannot be sure that the specific factors we chose to express it are workable. That is the proper subject for back-testing.

We also need to consider holding periods. I initially assumed the screen would be refreshed and the portfolios would be reconstituted every four weeks. That’s a time frame that often works for me. I’ve also been stretching things out lately and working with 13-week refresh periods but this strategy couldn’t even come close to handling that. I did, however, test it with 1- and 4-week refresh periods.

Tables 7, 8 and 9 show the results of tests run over 5- 10- and 1-year time horizons respectively with 4- and 1-week refresh periods.

Table 7:  5 Year Test

	4-week Hold Periods		1-week Hold Periods
	Micro-Nano	Benchmark	Micro-Nano	Benchmark
	Stock Model	IWC	Stock Model	IWC
Annual % Return	22.75%	13.08%	31.05%	13.08%
Standard Dev.	19.00%	15.58%	19.57%	15.58%
Max Drawdown	-28.31%	-28.96%	-29.87%	-28.96%
Beta	0.77		0.86
Annual Alpha %	14.09%		20.35%
Avg. 4 wk periods
All	1.59%	0.96%	0.35%	0.20%
Up markets	4.28%	3.84%	1.83%	1.73%
Down Markets	-2.51%	-3.41%	-1.52%	-1.73%

Table 8:  10 Year Test

	4-week Hold Periods		1-week Hold Periods
	Micro-Nano	Benchmark	Micro-Nano	Benchmark
	Stock Model	IWC	Stock Model	IWC
Annual % Return	25.09%	9.65%	30.21%	9.65%
Standard Dev.	27.52%	21.67%	26.46%	21.67%
Max Drawdown	-57.70%	-55.81%	-53.32%	-55.81%
Beta	1.02		1.01
Annual Alpha %	15.44%		20.51%
Avg. 4 wk periods
All	1.78%	0.65%	0.41%	0.21%
Up markets	5.71%	3.45%	2.42%	2.26%
Down Markets	-4.18%	-4.75%	-2.12%	-2.36%

Table 9:  1 Year Test

	4-week Hold Periods		1-week Hold Periods
	Micro-Nano	Benchmark	Micro-Nano	Benchmark
	Stock Model	IWC	Stock Model	IWC
Annual % Return	11.28%	23.24%	26.35%	23.24%
Standard Dev.	16.61%	12.30%	17.16%	12.30%
Max Drawdown	-15.30%	-9.96%	-13.82%	-9.96%
Beta	0.99		0.98
Annual Alpha %	-7.53%		6.23%
Avg. 4 wk periods
All	0.63%	1.62%	0.27%	0.45%
Up markets	3.39%	3.91%	1.70%	1.78%
Down Markets	-4.87%	-2.95%	-1.47%	-1.17%

Step Four – Assessing the Strategy

We’ve added a lot to the disastrous naive own-all-the-small-fry strategy whose results were shown in Table 6, enough so to allow us to say the full blown VQS version of the strategy works. Note, though, that it was cold as ice in the last year.

The last 12 months of weakness do not invalidate the theoretically sound and predominantly well-tested model. But they do force us to revisit the theoretical considerations that gave rise to the approach (something we should always be doing anyway) and work to understand the reasons for the recent frostiness.

The economy has been and still looks fine so I doubt pessimism about growth prospects is the problem. But with so many talking so much about interest rates likely to rise, who will be hurt and helped by the Administration’s tax and trade policies, etc., it may well be that the market’s been gently and quietly tip toeing toward the proverbial risk-off stance by de-emphasizing the riskiest subpart of the riskiest part of the market. Whether you want to fiddle with these stocks now, or watch from the sidelines, is a personal choice. I don’t care what choice you make, as long as it’s a knowing choice.

Conclusion

This VQS Micro-Nano strategy is investable, especially as presented here, as an idea generator rather than an automated execute-all-the-trades model. Unlike most models I come up with, it’s OK to refresh this one every week. Portfolio123 users who want to copy and work with this screen (and of course amend it if they wish) can access it by clicking here (it has been set for public visibility).

 

I’ll update the list less aggressively on this site.

Here are the stocks that currently pass muster in order, from high to low, of their VQS rank scores:

Entravision Communications Corp. (EVC)

Finjan Holdings Inc (FNJN)

SORL Auto Parts Inc (SORL)

DHI Group Inc (DHX)

Commercial Vehicle Group Inc (CVGI)

New York & Co Inc (NWY)

Palatin Technologies Inc. (PTN)

J Jill Inc (JILL)

Arotech Corp (ARTX)

AXT Inc (AXTI)

Audiocodes Ltd (AUDC)

North American Construction Group Ltd (NOA)

Orion Group Holdings Inc (ORN)

U.S. Auto Parts Network Inc (PRTS)

Performant Financial Corp (PFMT)

Amtech Systems Inc (ASYS)

Dynex Capital Inc. (DX)

Smart Sand Inc (SND)

R.R. Donnelley & Sons Co (RRD)

Nova Lifestyle Inc (NVFY)