The BFCM Trap: Why We're All Prisoners in a Game We Know We'll Lose

I work in ecommerce. November is when I wish I'd studied economics properly instead of just reading about it obsessively in my spare time. Because every year, I watch an entire industry engage in what can only be described as coordinated margin destruction, and every year we tell ourselves it's necessary because everyone else is doing it.

But here's the thing: it's not just retail madness. It's a perfectly predictable equilibrium outcome of a multi-period pricing game with strategic consumers and competitive pressure. The question isn't why Black Friday/Cyber Monday exists. It's why we persist in playing a game where everyone loses, despite knowing exactly what we're doing.

Let me show you the math. Fair warning: I'm an amateur economist with a dangerous amount of free time and access to game theory textbooks. What follows is my attempt to formalize what every ecommerce operator feels in their bones every November.

The Consumer's Problem: To Wait or Not to Wait

Consider a consumer deciding whether to buy something now or wait for BFCM. Let's formalize this.

The consumer has utility function U(c, m) where c is consumption and m is money remaining. They face two periods: regular season (period 0) and BFCM (period 1).

Regular price is p₀. Expected BFCM price is p₁ = p₀(1 - d) where d is the discount rate.

The consumer's decision problem is:

max[t ∈ {0,1}] U(ct, m - pt) - δᵗC(t)

where δ is the discount factor (how much they value future consumption relative to present consumption) and C(t) is the cost of waiting.

That waiting cost matters enormously and includes:

• Not having the thing they want right now

• Risk the item sells out

• Having to remember to check prices during BFCM

• Monitoring multiple retailers to find the best deal

• The psychological cost of uncertainty

  
  

The consumer buys in period 0 if and only if:

U(c₀, m - p₀) ≥ δ[U(c₁, m - p₁) - C(1)]

For a risk-neutral consumer (which most aren't, but bear with me), this simplifies beautifully to:

p₀ - p₁ ≥ δC(1)

Or, substituting our discount rate:

d · p₀ ≥ δC(1)

This gives us the critical discount rate d* where the consumer is perfectly indifferent:

d* = δC(1) / p₀

Translation: Strategic waiting is optimal when the discount exceeds this critical rate. The threshold increases with waiting costs and decreases with the initial price.

This is why expensive items see higher wait rates. A 20% discount on a $1,000 television is $200. A 20% discount on a $20 phone case is $4. The absolute savings matter, not just the percentage.

The Segmentation Problem

Here's where it gets interesting. Not all consumers have the same waiting costs.

Let C(1) be distributed according to some probability distribution F(c). Some people have very low costs of waiting (students with time and no money). Some have very high costs (busy professionals who value convenience).

The fraction of consumers who wait is:

λ(d) = F(d · p₀ / δ)

This creates natural market segmentation:

• High waiting cost consumers: buy at regular price

• Low waiting cost consumers: wait for BFCM
  

And this, friends, is the textbook condition for profitable price discrimination. You're serving two different customer types at two different price points. In theory, this should be great for retailers.

In theory.

The Retailer's Pricing Problem

Consider a single retailer with marginal cost c facing two consumer types: impatient with valuation θH (mass μ) and patient with valuation θL (mass 1-μ).

Assume θH > θL > c, and patient consumers will wait if they expect a discount.

The retailer wants to maximize:

max[p₀, p₁] μ(p₀ - c) + δ(1-μ)(p₁ - c)

But they face incentive compatibility constraints. High-valuation consumers shouldn't want to wait:

θH - p₀ ≥ δ(θH - p₁) [ICH]

Low-valuation consumers shouldn't want to buy early:

θL - p₁ ≥ (1/δ)(θL - p₀) [ICL]

The optimal strategy involves p₀ > p₁, with the discount chosen to make high-valuation consumers just barely indifferent to waiting.

Setting ICH to bind:

p₀ = θH - δ(θH - p₁)

Rearranging:

p₁ = θH - (θH - p₀)/δ

This is the fundamental relationship. The period 1 price is determined by how much you can extract from high-value customers in period 0 without making them wait.

The Credibility Problem (Or: Why BFCM Discounts Must Be Real)

Here's where my amateur economist brain really lights up, because this is where theory meets the reality I see every day.

The retailer's optimal strategy after period 0 is over, taking period 0 sales as given, might be different from what they promised. If consumers are sophisticated (and they increasingly are), they anticipate this.

The BFCM discount is only credible if consumers believe you'll actually follow through. And here's the kicker: the discount must be deep enough to be credible.

If you announce 10% off but consumers know you could profitably go deeper, they don't believe you. They keep waiting. Your announced sale doesn't clear the market.

This is why "up to 70% off" language exists. It's not just marketing puffery. It's a credible signal that some deals will be substantial enough to be worth the wait.

Competition and the Prisoner's Dilemma (The Part Where Everything Goes Wrong)

Now introduce competition. Two retailers, identical costs, facing the same pool of consumers who will buy from whoever offers the lowest price in each period.

Each retailer chooses (pi0, pi1). The payoff for retailer i is:

πi = si0(pi0 - c) + δsi1(pi1 - c) where s-it is market share in period t.

Standard Bertrand competition logic suggests that in period 1, both retailers end up pricing at marginal cost: pi1 = p-i,1 = c.

This means zero profit on BFCM sales.

The only profit comes from period 0 sales to impatient consumers. And as consumers learn this pattern, more of them become patient. The pool of period 0 sales shrinks. Margins compress.

This is the prisoner's dilemma in its purest form:

• If both retailers could credibly commit to modest discounts, both would be better off

• But given that the other retailer will discount heavily, your best response is to discount heavily too

• The Nash equilibrium is mutual defection (deep discounts) even though mutual cooperation (modest discounts) would be better for everyone
  

Formally, if both choose low discounts: profits are πH each

If both choose high discounts: profits are πL each

If one deviates: the high-discount retailer captures all delayed demand and earns πM

Where πM > πH > πL > 0, and the unique Nash equilibrium is (high, high) even though (low, low) would make everyone better off.

This is why BFCM exists and persists despite being collectively irrational.

The Expectations Spiral

It gets worse. Consumer expectations adapt based on observed history.

Let d̂t be consumers' expectation in period t of the discount in period t+1. If consumers update beliefs adaptively:

d̂t = α · d̂(t-1) + (1-α) · d(t-1)

where α ∈ (0,1) is the weight on prior expectations.

When consumers expect deep discounts, they wait. When they wait, period 0 revenue falls. When period 0 revenue falls, pressure increases to discount in period 1. When discounts deepen in period 1, expectations for future discounts increase.

This is a ratchet effect. Once discount expectations are high, they're extraordinarily difficult to reverse. You can't just announce "we're doing smaller discounts this year" because consumers won't believe you until they see it, and you can't afford to be the first mover.

What The Data Actually Shows

I don't have access to comprehensive industry data, but the model generates testable predictions that align with what I observe:

On consumer behavior:

• Waiting rates increase with product price (higher absolute savings)

• High-income consumers purchase at regular prices more often (higher opportunity cost of time)

• Search intensity spikes in weeks before BFCM (updating expectations)

On retailer behavior:

• Competitive markets show deeper discounts (more Bertrand-like competition)

• Multi-product retailers use BFCM as loss leaders on popular items while maintaining margins on complements

• Categories with fewer competitors sustain shallower discounts

On dynamics:

• Discount depth has increased over time (expectations ratchet)

• BFCM window has expanded (retailers trying to capture early buyers before full competition hits)

• Stockout rates have increased (strategic undersupply attempting to support higher prices)

  
  

The Welfare Question Nobody Asks

Who actually benefits from BFCM?

Consumer surplus under price discrimination is:

CS = ∫[0 to μ](θH - p₀) + ∫[μ to 1]δ(θL - p₁)

Compared to a single-price regime:

CS(SP) = ∫[0 to 1]max(0, θi - p*)

Patient consumers benefit from lower BFCM prices. Impatient consumers lose from higher regular prices. Whether total consumer surplus increases depends on the distribution of consumer types and the depth of competition.

Under perfect competition in period 1 (which BFCM approaches), producer surplus reduces to:

PS = μ(p₀ - c)

Retailers are worse off than they would be with coordinated modest discounts, but better off than with year-round Bertrand competition.

The efficiency question is whether the gains from better matching (different prices for different consumer types) exceed the losses from:

• Increased waiting costs

• Operational chaos (everyone trying to fulfill orders simultaneously)

• Pulled-forward demand (people buying in November who would have bought in December anyway)

• Strategic behavior costs (time spent monitoring prices, comparison shopping)

  
  

My suspicion, based on nothing but spidey suspicion and looking at things sideways, is that total welfare decreases. The matching gains are modest. The operational and strategic costs are substantial.

Why We Can't Escape

If the equilibrium is suboptimal for everyone, why can't we just... stop?

Because coordination is impossible without enforcement. Any explicit agreement to limit discounting would violate antitrust law. Implicit coordination fails because the incentive to defect is too strong.

The game theory is unambiguous: if you believe your competitor will offer 40% off, your best response is to offer 45% off. If they believe you'll offer 45% off, their best response is 50% off. We race to the bottom of marginal cost.

There are theoretical escape routes:

Commitment mechanisms: Price matching guarantees, public statements, loyalty programs that create switching costs. But these are weak and hard to maintain.

Product differentiation: If products aren't perfect substitutes, competition softens. This is why private label and exclusive partnerships matter more during BFCM than other times.

Supply constraints: Strategic undersupply can support higher prices, but requires credible commitment not to expand supply ex post. Consumers see through this quickly.

Category expansion: Instead of discounting existing inventory, introduce new products exclusively for BFCM. This avoids cannibalization but requires substantial product development investment.

Temporal expansion: Spread "BFCM" across weeks rather than days. This smooths operations and reduces competitive pressure at any single point. We're seeing this happen organically, which is probably the only equilibrium-shifting change that's actually working.

What This Means If You Actually Run A Business

The formal analysis reveals why individual optimization leads to collective suboptimization. Your best response is to discount given that others discount. The equilibrium is stable but inefficient.

You cannot unilaterally exit BFCM without losing substantial market share. You cannot coordinate with competitors without violating law. You cannot credibly commit to future restraint because consumers have learned not to believe such commitments.

What you can do:

Measure your actual incremental lift: Most retailers don't know if BFCM sales are truly incremental or just pulled forward from December. If it's mostly pulled forward, you're destroying margins for no gain in annual revenue.

Optimize your discount structure: Not all discounts are equal. Loss leaders on high-visibility items plus maintained margins on complements can work better than across-the-board percentage cuts.

Invest in operational efficiency: If you're going to play the game, play it well. The winner in a race to marginal cost is whoever has the lowest marginal cost.

Build switching costs: Loyalty programs, subscriptions, proprietary ecosystems. Anything that makes your customers less likely to price-shop with competitors.

Accept the equilibrium: Sometimes the best strategy is to acknowledge you're in a prisoner's dilemma and optimize within constraints rather than fighting the inevitable.

The Depressing Conclusion

BFCM is a Nash equilibrium. It's stable. It's collectively irrational. And we're all trapped in it.

The mathematics are clear. The incentives are clear. The outcome is predictable. And yet every November, we wake up, look at our margins, and do it again.*

Because the alternative is being the one retailer who doesn't participate while your competitors capture all the price-sensitive demand.

This is economics in action. Not the economics of optimal allocation and efficient markets. The economics of coordination failures, strategic behavior, and game-theoretic traps we can see but cannot escape.

I work in ecommerce. I understand the trap. And next November, I'll participate in it anyway.

Because what else can I do? *Written by someone who reads too much game theory and has to explain to the C-Suite every year why BFCM margins are terrible despite everyone knowing in advance that BFCM margins will be terrible.