Case Study 16.2 — The Cellar as an Emergency Asset: What the 2020 Shutdowns Revealed About Wine Inventory

Type: real public event, with a clearly-labeled composite operator built from documented industry patterns Relevant sections: §16.6 (capital tied up in bottles), §16.8 (pruning), §16.3 (three-tier) Sourcing: Tier 1 for the COVID-19 dining-room closures of spring 2020 and the widely reported emergency measures that permitted many restaurants to sell sealed alcohol for off-premise consumption. Tier 2 for characterizations of how operators responded. Tier 3 — clearly constructed — for the composite restaurant's figures, which are illustrative and do not describe any real business.

Background: the one asset that could be sold

In March 2020, dining rooms across the United States closed, in most places by order and nearly everywhere within a few days of one another. Restaurants that had been operating normally on a Friday had, by the following week, no on-premise revenue at all.

What followed is well documented: a rapid, improvised expansion of off-premise service, third-party delivery volume that spiked and then stayed elevated, and — the piece that matters here — emergency regulatory measures in many states permitting restaurants to sell sealed alcohol for off-premise consumption, something that in normal times would have violated the terms of an on-premise license.

Those measures were not uniform. Some states permitted sealed wine and beer only. Some permitted mixed drinks in sealed containers. Some permitted the sale of a restaurant's existing inventory at retail. Some permitted nothing at all. Several states have since made some version permanent; several let the authority lapse; the question remained contested in state legislatures for years afterward.

For this chapter's purposes, one consequence stands out. When on-premise revenue went to zero, most of what a restaurant owned could not be converted to cash. The hearth could not be sold. The lease was a liability, not an asset. The walk-in's contents had days of life. The wine cellar was, for a great many restaurants, the single largest liquid asset in the building — and in the jurisdictions that allowed it, operators sold it.

The operating issue: what you learn about an asset when you have to sell it

The composite (clearly constructed; illustrative figures, not any real business)

A 110-seat independent in a large American city. Roughly \$2.6 million in annual sales, a well-known wine program, a list of about 280 selections, and a cellar the owners were proud of. Wine ran about 18% of revenue, which is a serious program.

Inventory at cost, in early March: about \$140,000.

When the dining room closed, the operators did the arithmetic every restaurant did that month. Two weeks of payroll for a skeleton crew, rent, insurance, and the utility bills on a building that had to keep its coolers running: call it \$60,000 a month of unavoidable outflow with essentially no revenue. Their cash on hand was roughly six weeks of that.

Their state permitted retail sale of existing inventory. They opened the cellar.

Three things happened, and all three are lessons this chapter has been building toward.

One: it sold, but not at cost-plus. Selling wine to the public is a retail transaction, and retail buyers price-check on a phone. Wine that had been priced at \$95 on the list moved at something closer to retail — meaning the operators recovered their wholesale cost and a modest margin on the fast-moving middle of the cellar, and considerably less than that on the top.

Two: the fast shelf sold and the slow shelf did not. The same wines that turned nine or ten times a year on the list were gone in ten days. The prestigious bottles — the ones that had justified the cellar's reputation, the ones a wine director had chased for years — sat, because the small number of people who would pay for them were exactly the people who already owned them.

Three: the money that came out was less than the money that went in. Not catastrophically less. But an operator who had been carrying \$140,000 of inventory as a proud asset discovered that its cash value under pressure was a good deal lower than its book value, and that the gap was concentrated in precisely the part of the cellar that had cost the most to build.

What it shows

Strip away the pandemic and this is a general truth about inventory that most restaurant operators never have occasion to test.

Wine is capital, and capital has a liquidation value that is not its cost. Chapter 16 §16.6 computes Bellwether's inventory at \$4,068 and its turns at 11.4 a year. What that section does not say — and this case does — is that the \$780 sitting on the \$96–\$135 shelf is worth \$780 only in a world where you have time. Under pressure, the fast shelf is money and the slow shelf is furniture.

Turn rate is a liquidity measure, not just an efficiency measure. A shelf that turns 9.1 times a year converts to cash in about forty days by simply operating normally. A shelf that turns 2.7 times takes four and a half months. In a cash crisis those are not two versions of the same asset. One is nearly cash. The other is nearly not.

The regulatory structure decided who had the option at all. This is the sharpest point in the case and it links directly to Case Study 16.1. Two restaurants with identical cellars, in two different states, had entirely different balance sheets in April 2020 — because one state's emergency order permitted retail sale of existing inventory and the other's did not. The value of your wine inventory in a crisis is partly a function of a state law you did not write. No operator's projections had ever contained that variable.

And the concentration was the problem, not the total. The composite's \$140,000 was not irresponsible for a \$2.6 million restaurant with an 18% wine program. What made it painful was where it sat: heavily weighted toward selections that moved slowly by design, because a great cellar is supposed to have depth. The very thing that made the program distinctive made the asset illiquid.

Outcome and the contested decision

Many of the emergency alcohol measures were extended; a number were made permanent; others expired. The debate that followed — in state legislatures, among wholesalers, among restaurant associations — is itself instructive, because it was fought over the three-tier structure. Permanent off-premise alcohol sales by restaurants blur a line between tier three's on-premise and off-premise categories that most states had maintained since 1933.

Reasonable people disagreed. Restaurant associations generally argued the revenue was material and the public had adapted. Some wholesale and retail interests argued the categories exist for reasons and that a restaurant selling sealed wine to go is functionally a liquor store operating under a cheaper license.

A restaurant operator's honest position on this is that they do not control it. Which is exactly why the planning lesson matters.

The lesson

Build a wine list you could sell in a hurry.

Not literally — Bellwether is not planning for a shutdown, and no plan should be built around one. But the discipline the question imposes is healthy and available every day:

• Know your turn rate by shelf, not just in aggregate. The aggregate hides everything (§16.6).
• Cap the slow shelf deliberately. §16.8's rule — no more than 10% of selections and 12% of inventory dollars above the "sells regularly" ceiling — is a liquidity rule wearing a merchandising costume.
• Prefer breadth to depth at the top. One bottle anchors a price point as well as three and costs a third as much to own. This is true in ordinary times and decisive in bad ones.
• Remember that the cash you didn't spend is also an asset, and it is the only one with a liquidation value of exactly one hundred percent.

The composite's operators, asked afterward what they would do differently, gave an answer this chapter would endorse: they would have run the same list with half the depth, and kept the difference in the bank.

Discussion questions

1. Bellwether holds \$4,068 of wine inventory, of which \$780 sits on a shelf turning 2.7 times a year. Estimate what that \$780 would realize in a forced thirty-day sale, state your assumptions, and say what the exercise tells you about how deep to stock it.

2. This case argues that turn rate is a liquidity measure and not only an efficiency measure. Explain the difference, and name one other restaurant inventory category where the same distinction would apply.

3. The composite's \$140,000 cellar was not irresponsible for its size of business. What was the error, and at what point in the cellar's construction would you have wanted somebody to raise it?

4. Two restaurants with identical inventories had different balance sheets in April 2020 because of state law. What, if anything, can an operator do in advance about a variable like that? Is "nothing" an acceptable answer?

5. Argue the wholesaler's side of the permanent-off-premise-alcohol debate as strongly as you can. Then argue the restaurant association's side. Which one would you find more persuasive if you did not own a restaurant?

6. Case Study 16.1 argued that the three-tier system decides what you can buy and what you pay. This case argues it can also decide what your inventory is worth. Combine the two into a single paragraph you would put in a business plan's risk section.

7. §16.8 recommends deciding a bottle's exit before you buy it. Reread the four exits and say which of them would have been available to the composite operator in April 2020, and which would not.

8. The chapter's closing recommendation is "run the same list with half the depth and keep the difference in the bank." Under what circumstances would that advice be wrong?