Gran­ite City Food & Brew­er­y is the lat­est small pub­lic com­pany think­ing a­bout leav­ing its share­hold­ers in the dark.

That means that Gran­ite City would deregister its shares and get out of the costs of complying and disclosing, a task that can easily exceed $500,000 a year for even small companies. One of the first expensive things Granite City won’t have to do is file the big annual report that the Securities and Exchange Commission requires every year.

But no public filings means no readily available information about the company. That lack of disclosure is one rea­son why the whole proc­ess of deregistering is called “going dark.”

Going dark is mostly a small-company strategy, and there are a number of dark companies around our region. Like Gran­ite City, they got tired of pay­ing to comply with SEC rules, and their stocks were not actively trading anyway.

But of all the op­tions, going dark seems eas­i­ly to be the worst. Share­hold­ers already own a relatively inactive stock that isn’t easy to sell at a fair valuation. Being left in the dark on how the business is performing financially just makes that a whole lot worse. Af­ter all, the rea­son we have an SEC is that we have agreed that fair­ness and trans­par­en­cy help cre­ate an ef­fi­cient cap­i­tal mar­ket.

Instead of lingering on the stock market with little public access to company performance, it would be far better to find a buyer for all the company shares, or have insiders and big holders become the buyer. At least try.

That’s what Gran­ite City ought to be doing, though it didn’t return a call seeking to discuss the matter. A pri­vate equi­ty firm from Dal­las is the controlling shareholder. The firm should fig­ure out a fair price to pay the oth­er own­ers for their Gran­ite City shares and buy them.

Granite City isn’t a start-up anymore, with about $34 million in revenue for its last reported quarter, but it is small for the public markets. In its late October announcement about possibly going dark, the company said “in light of the company’s size, small market capitalization and the thinly traded market for its stock,” the “financial burden of reporting [may be] disproportionate to any benefits.”

That idea, what Dor­sey & Whit­ney part­ner Ted Far­ris called a “bal­an­cing of in­ter­est,” is the legal principle behind a going-dark process. Yes, it doesn’t help investors trade when their company goes dark, but they do get the ben­e­fit of lower costs.

“In the case of Gran­ite City, they had al­read­y been de­listed by Nasdaq,” Far­ris said. “The com­pany no long­er has the ben­e­fit of the list­ing, and but it still has all the costs and legal re­port­ing ob­li­ga­tions a list­ed com­pany would have.”

But then a­gain, the high costs for be­ing pub­lic borne by too small of a com­pany like Gran­ite City would have been a great dis­cus­sion to have had back in June 2000. That is, one day be­fore the com­pany went a­head with its in­itial pub­lic of­fer­ing. That IPO net­ted about $3.5 mil­lion and came to mar­ket less than a year af­ter the com­pany had op­ened its first and then-only res­tau­rant in St. Cloud.

Yes, times were dif­fer­ent then, but the sit­u­a­tion in 2013 for Gran­ite City could also have been eas­i­ly pre­dict­ed.

An­oth­er company that nev­er should have been pub­lic is Za­re­ba Systems of Ply­mouth. Not a single Zareba share traded on Nasdaq 36 of the 50 trading days prior to Zareba deciding in the sum­mer of 2009 to go dark. Yet Zareba’s is a curious case of more or less accidentally doing the right thing, because it ended up being sold rather than going through with the deregistration.

Dale Nordquist, who was CEO at the time, said the company was “going dark to fa­cili­tate a stra­tegic sale,” and a com­pet­i­tor named Wood­stream Corp. “mis­un­der­stood our in­ten­tion,” as Wood­stream did not wait for that sale proc­ess to gear up and prompt­ly re­newed its ef­fort to ac­quire Za­re­ba.

Hunt Greene, an investment banker who ad­vised Za­re­ba, said Nordquist’s ex­pla­na­tion for going dark was per­haps a post-deal ra­tion­al­i­za­tion, but both a­gree that it led to a great out­come. Penn­syl­vania-based Wood­stream paid more than four times what the company was valued on Nasdaq the day before the going-dark idea was announced.

Greene said company boards that are serious about finding a good buyer can skip deregistering the shares, and instead publicly announce the search for a good buyer. That gives competitors who had been thinking about making a bid a sense of urgency about making their move.

Not ev­er­y­bod­y agrees that trying to find a buyer beats going dark for reasons that are maybe as simple as the directors can’t imagine anyone who would want to buy their company. In any case, we now have the odd sit­u­a­tion of watch­ing active trad­ing in the stock of com­panies that have long been dark.

Even with up-to-date fil­ings by pub­lic­ly held com­panies, buy­ing stock in them is one of the ex­am­ples an eco­nom­ics pro­fes­sor might use when dis­cuss­ing the prob­lem of one side of a trans­ac­tion knowing a whole lot more a­bout what some­thing is re­al­ly worth than the oth­er side does.

Going dark clear­ly can make that a lot worse.

There are dark com­panies that do try to inform holders. Mark Thom­as, the CEO of HEI Inc. of Vic­to­ri­a, sev­er­al times re­ferred to HEI as “pub­lic” in a brief con­ver­sa­tion, even though it went dark at the end of 2007. That’s be­cause HEI puts audit­ed fi­nan­cial state­ments and quarterly an­nounce­ments of re­sults on its website.

Delphax Technologies, a Bloomington-based pro­duc­er of print­ers and sup­plies that went dark in late 2008, has a dif­fer­ent ap­proach. Back when I was work­ing with Delphax 12 years ago, in­for­ma­tion was read­i­ly avail­able, but in check­ing last week, it looked like the most-re­cent fi­nan­cial news is just a summary with no balance sheet information — from the September 2011 fiscal year.

That did not stop shares of Delphax stock from chan­ging hands last week, last at 45 cents each.

What that last trade was based on is an­y­one’s guess, but the good news is that the prob­lem of unequal information probably wasn’t an is­sue.

With both sides in the dark, nei­ther one may have had any i­de­a what was going on.