
Econ Market Structure
- 29 installs
- 223 repo stars
- Updated June 6, 2026
- asgard-ai-platform/skills
econ-market-structure is a skill that classifies markets across the four competitive structures to predict firm pricing behavior and market outcomes.
About
This skill analyzes market structures (perfect competition, monopolistic competition, oligopoly, monopoly) to predict firm behavior and market outcomes. A developer or strategist uses it to classify a market's competitive structure, predict pricing, or understand why an industry behaves as it does. It classifies by firm count, differentiation and barriers, then predicts behavior and policy implications.
- Classifies markets across perfect competition, monopolistic competition, oligopoly and monopoly
- Predicts pricing and long-run profit behavior from structural characteristics
- Flags antitrust and contestable-market implications
Econ Market Structure by the numbers
- 29 all-time installs (skills.sh)
- Ranked #1,873 of 3,282 Productivity & Planning skills by installs in the Skillselion catalog
- Data as of Aug 2, 2026 (Skillselion catalog sync)
econ-market-structure capabilities & compatibility
- Capabilities
- market classification · competitive analysis · pricing prediction
- Use cases
- research · planning
What econ-market-structure says it does
Market structure determines how firms compete, set prices, and earn profits.
IRON LAW: Structure Determines Behavior, Not Vice Versa
npx skills add https://github.com/asgard-ai-platform/skills --skill econ-market-structureAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 29 |
|---|---|
| repo stars | ★ 223 |
| Last updated | June 6, 2026 |
| Repository | asgard-ai-platform/skills ↗ |
What it does
Classify a market's competitive structure to predict pricing and firm behavior.
Who is it for?
Classifying whether a market is competitive, oligopolistic or a monopoly and predicting its pricing behavior.
Skip if: Calling a firm a monopoly just because it has a unique product; structure is about the whole market.
When should I use this skill?
The user asks why a firm can charge so much, whether a market is competitive, or if prices will come down.
What you get
Produces a market-structure classification with evidence, predicted behavior and policy implications.
- market-structure classification
- structural evidence
- predicted behavior
By the numbers
- 4 market structures
- 4 classification steps
Files
Market Structure Analysis
Overview
Market structure determines how firms compete, set prices, and earn profits. The four structures — perfect competition, monopolistic competition, oligopoly, monopoly — predict increasingly different behaviors as concentration rises and differentiation increases.
Framework
IRON LAW: Structure Determines Behavior, Not Vice Versa
Classify the market structure FIRST based on structural characteristics
(number of firms, barriers, differentiation), THEN predict behavior.
"This company charges high prices" does not mean it's a monopoly —
high prices can occur in oligopolies and even monopolistic competition.The Four Structures
| Feature | Perfect Competition | Monopolistic Competition | Oligopoly | Monopoly |
|---|---|---|---|---|
| Firms | Very many | Many | Few | One |
| Product | Homogeneous | Differentiated | Homogeneous or differentiated | Unique, no close substitutes |
| Entry barriers | None | Low | High | Very high |
| Price power | None (price taker) | Some (limited by substitutes) | Significant (interdependent) | Full (price maker) |
| Long-run profit | Zero (economic) | Zero (economic) | Positive possible | Positive |
| Examples | Agricultural commodities, forex | Restaurants, clothing | Airlines, telecom, auto | Utilities, patents |
Classification Steps
1. Count sellers: How many significant firms serve this market? 2. Assess differentiation: Are products identical or differentiated? 3. Evaluate entry barriers: Can new firms enter easily? 4. Check interdependence: Do firms react to each other's moves?
Behavior Predictions by Structure
Perfect Competition: Price = marginal cost. Firms are price takers. No advertising needed. Long-run economic profit = 0.
Monopolistic Competition: Short-run profits possible through differentiation. Long-run: entry erodes profits to zero. Firms compete on brand, quality, location.
Oligopoly: Firms are interdependent — each watches rivals' moves. Game theory applies. May collude (tacitly or explicitly). Kinked demand curve or Cournot/Bertrand models.
Monopoly: Price > marginal cost. Deadweight loss exists. May be regulated (utilities) or temporary (patents). Natural monopolies occur when average costs decline with scale.
Output Format
# Market Structure Analysis: {Industry}
## Classification
- Structure: {type}
- Evidence:
- Number of firms: ...
- Product differentiation: ...
- Entry barriers: ...
- Interdependence: ...
## Predicted Behavior
- Pricing: {price-taking / markup / strategic}
- Long-run profit: {zero / positive}
- Competition type: {price / quality / advertising / innovation}
## Policy Implications
{Antitrust concerns, regulation needs, consumer impact}Examples
Correct Application
Scenario: Taiwan's telecom market
- Firms: 3 major (中華電信, 台灣大, 遠傳) + 2 minor → Few
- Differentiation: Moderate (speed/coverage differences, but largely substitutable)
- Entry barriers: Very high (spectrum licenses, infrastructure costs ~NT$100B+)
- Interdependence: High (price changes by one trigger immediate responses)
- Classification: Oligopoly ✓
- Predicted behavior: Tacit price coordination, competition on bundling and service rather than price, stable high margins
Incorrect Application
- "iPhone has no competitors so Apple is a monopoly" → Smartphones have many competitors (Samsung, Google, Xiaomi). Apple has market power through differentiation, but the smartphone market is oligopoly, not monopoly. Structure is about the market, not one firm's product uniqueness.
Gotchas
- Market definition changes the answer: "Smartphones" is an oligopoly. "iOS devices" is a monopoly. The market boundary determines the structure classification.
- Perfect competition is theoretical: Almost no real market is perfectly competitive. Use it as a benchmark, not a classification for real industries.
- Oligopoly is the most complex: Game theory, collusion, and strategic behavior make oligopoly analysis harder than other structures. Be explicit about assumptions.
- Contestable markets: Even a monopoly may behave competitively if entry barriers are low (threat of entry disciplines pricing). Barriers matter as much as current firm count.
- Dynamic markets: Tech markets may look like monopolies today (Google Search) but face competitive pressure from disruption (AI chat). Consider trajectory, not just snapshot.
References
- For oligopoly game theory models (Cournot, Bertrand, Stackelberg), see
references/oligopoly-models.md
Example: 台灣外送平台市場結構分析
Scenario
一位新創顧問正在評估是否協助客戶進入台灣餐飲外送市場。她的問題是:
「foodpanda 和 Uber Eats 兩家幾乎壟斷了台灣外送市場,但它們的抽成還是高達 30%,餐廳都在抱怨。請問這個市場到底是什麼結構?為什麼價格下不來?未來有機會打進去嗎?」
---
Analysis
Step 1:計算賣方數量
台灣外送平台現狀(2025 年):
| 平台 | 市占率(訂單量估計) | 狀態 |
|---|---|---|
| foodpanda | ~52% | 活躍,Delivery Hero 旗下 |
| Uber Eats | ~44% | 活躍,Uber 旗下 |
| 其他(murmur、在地小平台) | ~4% | 邊緣,多已退出 |
結論:顯著競爭者 2 家,高度集中。
---
Step 2:評估產品差異化程度
- 外送速度、介面體驗、餐廳覆蓋略有差異,但消費者對兩平台普遍有帳號且習慣交叉使用
- 餐廳端:兩平台合約條款、抽成結構(28–32%)高度相似
- 消費者端:配送費、折扣券差異存在,但核心服務同質性高
結論:輕度差異化(near-homogeneous,偶有促銷差異)
---
Step 3:評估進入障礙
| 障礙類型 | 具體內容 |
|---|---|
| 網絡效應 | 餐廳多 → 消費者多 → 外送員多,三邊市場形成強大護城河 |
| 資本門檻 | 外送員補貼、行銷費用、科技基礎設施:初期燒錢規模 $10 億新台幣以上 |
| 品牌信任 | 消費者熟悉介面,轉換成本低但習慣黏著高 |
| 規模經濟 | 訂單密度越高,單筆配送成本越低;新進者在密度稀疏時無法獲利 |
結論:進入障礙 非常高(規模+網絡效應雙重鎖定)
---
Step 4:確認相互依賴性
- 2021 年 foodpanda 曾嘗試漲配送費,Uber Eats 未跟進 → foodpanda 隔月回調
- 兩平台抽成均維持在 30% 左右,從未發動「抽成戰」
- 促銷策略(折扣券、免運門檻)時常同步推出,顯示高度監視對手行為
結論:相互依賴性 極高,符合寡占行為特徵
---
市場結構判定:寡占(Duopoly)
四項指標全部指向 寡占,且更精確為雙寡頭(Duopoly)。
---
行為預測
定價:抽成維持高位(30%)並非巧合——這是寡占廠商避免破壞性價格戰的理性選擇。兩家都清楚,若降抽成將引發對手跟進,雙方均虧損卻市占不變。
競爭維度:競爭集中在消費者促銷(折扣券)和餐廳獨家合作,而非抽成率或定價本身。
長期利潤:兩家均有持續正利潤空間(或至少策略性容忍虧損以防對手擴張)。
新進者威脅評估:
- 即便燒錢進入,也面臨既有雙寡頭可以針對性補貼防禦
- 除非擁有差異化優勢(例如:特定垂直市場如生鮮、企業餐),否則正面競爭勝率極低
---
Result
Market Structure Analysis: 台灣餐飲外送平台
Classification
- Structure: 寡占(Duopoly)
- Evidence:
- Number of firms: 2 家主要競爭者(foodpanda 52%、Uber Eats 44%),合計市占 96%
- Product differentiation: 輕度差異化,核心服務同質,消費者與餐廳均難以完全仰賴單一平台
- Entry barriers: 極高——三邊網絡效應 + 高燒錢需求 + 規模密度門檻
- Interdependence: 極高——定價跟漲、促銷同步、避免抽成戰均為典型寡占行為
Predicted Behavior
- Pricing: 策略性定價(抽成鎖定 30%,以默契協調避免競爭)
- Long-run profit: 正利潤可持續,受新進者威脅程度低
- Competition type: 消費者端促銷競爭 + 餐廳端獨家合作競爭,非價格競爭
Policy Implications
- 抽成問題:30% 抽成對餐廳構成實質擠壓,但在雙寡頭結構下不會自然下調——若公平交易委員會認定存在默契協調,可啟動調查
- 新進者建議:正面進入整體外送市場風險極高。建議聚焦可防禦的垂直利基(如企業午餐訂餐、生鮮冷鏈配送),在該利基先建立密度優勢,再考慮橫向擴張
- 消費者福祉:目前市場結構產生明顯的消費者剩餘損失(高抽成轉嫁餐廳 → 菜單售價上漲),政策介入正當性存在,但需與平台基礎設施投資效益權衡
---
關鍵洞察:「兩家就是壟斷」是常見誤解。雙寡頭有競爭,但競爭是策略性的而非價格破壞性的。這正是寡占理論的核心預測:廠商會在不觸發對手報復的範圍內競爭,因此抽成「下不來」是結構決定行為的直接結果,不是廠商邪惡,而是理性均衡。
Oligopoly Models: Cournot, Bertrand, Stackelberg
寡占分析的核心問題:廠商如何在互依環境中做決策? 三個模型給出不同答案,因為它們對「競爭變數」的假設不同。
---
模型選擇指引
| 情境 | 適用模型 |
|---|---|
| 廠商同時決定產量,產品同質 | Cournot |
| 廠商同時決定價格,產品同質 | Bertrand |
| 廠商同時決定價格,產品有差異 | Bertrand 差異版 |
| 一家廠商先動、其他跟進 | Stackelberg |
| 廠商重複互動,可能默契合謀 | 重複賽局(見下文) |
選模型時先問:「廠商真正掌控的決策變數是什麼?」航空公司掌控班次(容量 ≈ 產量) → Cournot;零售商掌控標牌價格 → Bertrand。
---
Cournot 模型(數量競爭)
核心假設
- 兩家(或多家)廠商同時決定產量 $q_1, q_2$
- 產品同質,市場存在反需求函數 $P = a - b(q_1 + q_2)$
- 每家廠商將對手產量視為給定(Cournot 猜測,conjecture = 0)
推導兩家廠商的均衡
符號定義
| 符號 | 意義 |
|---|---|
| $a$ | 截距(最高願付價格) |
| $b$ | 需求斜率(正值) |
| $c$ | 邊際成本(兩廠商相同,簡化版) |
| $q_1, q_2$ | 廠商 1、2 的產量 |
廠商 1 的利潤
$$\pi_1 = P \cdot q_1 - c \cdot q_1 = [a - b(q_1 + q_2) - c] q_1$$
對 $q_1$ 取一階條件
$$\frac{\partial \pi_1}{\partial q_1} = a - 2bq_1 - bq_2 - c = 0$$
廠商 1 的反應函數(Best Response Function)
$$q_1^* = \frac{a - c}{2b} - \frac{q_2}{2}$$
對稱地,廠商 2:
$$q_2^* = \frac{a - c}{2b} - \frac{q_1}{2}$$
聯立求解 Nash 均衡
$$q_1^ = q_2^ = \frac{a - c}{3b}$$
均衡總產量、價格、利潤
$$Q^ = \frac{2(a-c)}{3b}, \quad P^ = \frac{a + 2c}{3}, \quad \pi_i^* = \frac{(a-c)^2}{9b}$$
數值範例
市場需求:$P = 100 - Q$(即 $a=100, b=1$),兩廠商邊際成本 $c = 10$
$$q_1^ = q_2^ = \frac{100 - 10}{3 \times 1} = 30$$
$$Q^ = 60, \quad P^ = 100 - 60 = 40$$
$$\pi_i^* = (40 - 10) \times 30 = 900$$
對照基準
| 情境 | 總產量 | 市場價格 | 廠商利潤(各) |
|---|---|---|---|
| 完全競爭($P=MC$) | 90 | 10 | 0 |
| Cournot 均衡 | 60 | 40 | 900 |
| 卡特爾(聯合壟斷) | 45 | 55 | 1012.5 |
| 獨佔 | 45 | 55 | 2025 |
Cournot 均衡介於完全競爭與壟斷之間。廠商數量 $n$ 增加時,Cournot 均衡趨近完全競爭。
N 家廠商的通式
$$q_i^ = \frac{a - c}{(n+1)b}, \quad P^ = \frac{a + nc}{n+1}$$
當 $n \to \infty$:$P^* \to c$(完全競爭)
---
Bertrand 模型(價格競爭)
核心假設
- 兩家廠商同時設定價格
- 產品同質,消費者全選最低價
- 廠商可以無限量供給(產能不受限)
Bertrand 悖論
在這些假設下,均衡是 $P_1 = P_2 = c$(邊際成本)。
為什麼?
- 若 $P_1 > P_2 > c$:廠商 1 失去所有客戶,有誘因降價至略低於 $P_2$
- 若 $P_1 = P_2 > c$:廠商 1 有誘因降一分錢奪全市場
- 唯一不想偏離的點:$P = c$
悖論之處:兩家廠商就產生「完全競爭」結果。現實中幾乎從不發生,因為假設太強。
Bertrand 悖論的解法
下列條件之一成立時,Bertrand 均衡不再是 $P = c$:
| 破解條件 | 機制 | 現實例子 |
|---|---|---|
| 產能限制 | 低價廠商無法供應全市場 | 航空座位有限 |
| 產品差異 | 消費者有品牌偏好,不純粹比價 | 連鎖餐廳 |
| 轉換成本 | 消費者不會為微小差價換供應商 | 銀行帳戶 |
| 重複賽局 | 低價今天獲利,但引發對手報復 | 長期市場關係 |
差異化 Bertrand 模型
產品差異化時,廠商 1 的需求:
$$q_1 = a - \alpha P_1 + \beta P_2 \quad (\alpha > \beta > 0)$$
廠商 1 的反應函數(對價格,非產量):
$$P_1^* = \frac{a + \alpha c}{2\alpha} + \frac{\beta}{2\alpha} P_2$$
注意:價格反應函數斜率為正(strategic complements)—— 對手漲價,我也跟漲。相較之下,Cournot 反應函數斜率為負(strategic substitutes)。
---
Stackelberg 模型(先後動)
核心假設
- 廠商 1(領導者,leader)先決定產量
- 廠商 2(追隨者,follower)後決定,且觀察到廠商 1 的選擇
- 使用逆向歸納(backward induction)求解
求解程序
Step 1:追隨者的反應函數(與 Cournot 相同)
$$q_2^* = \frac{a - c}{2b} - \frac{q_1}{2}$$
Step 2:領導者將追隨者反應代入自身利潤
$$\pi_1 = \left[a - b\left(q_1 + q_2^*(q_1)\right) - c\right] q_1$$
將 $q_2^* = \frac{a-c}{2b} - \frac{q_1}{2}$ 代入:
$$\pi_1 = \left[\frac{a-c}{2} - \frac{b q_1}{2}\right] q_1$$
Step 3:對 $q_1$ 最大化
$$\frac{\partial \pi_1}{\partial q_1} = \frac{a-c}{2} - b q_1 = 0 \implies q_1^* = \frac{a-c}{2b}$$
$$q_2^* = \frac{a-c}{4b}$$
數值範例(接續前例)
$P = 100 - Q, c = 10$
$$q_1^ = \frac{90}{2} = 45, \quad q_2^ = \frac{90}{4} = 22.5$$
$$Q^ = 67.5, \quad P^ = 32.5$$
$$\pi_1 = (32.5 - 10) \times 45 = 1012.5$$ $$\pi_2 = (32.5 - 10) \times 22.5 = 506.25$$
三模型比較(同一需求函數)
| 模型 | $q_1$ | $q_2$ | $Q$ | $P$ | $\pi_1$ | $\pi_2$ |
|---|---|---|---|---|---|---|
| Cournot | 30 | 30 | 60 | 40 | 900 | 900 |
| Stackelberg | 45 | 22.5 | 67.5 | 32.5 | 1012.5 | 506.25 |
| 卡特爾 | 22.5 | 22.5 | 45 | 55 | 1012.5 | 1012.5 |
先動優勢(First-Mover Advantage):Stackelberg 領導者比 Cournot 均衡多賺 112.5;追隨者比 Cournot 均衡少賺 393.75。先動優勢來自「承諾效果」——領導者透過可觀察的產量選擇,迫使追隨者縮減。
---
合謀與重複賽局
為何廠商有合謀誘因
在上例中,卡特爾利潤(各 1012.5)> Cournot 利潤(各 900)。廠商若能維持「各產 22.5」,雙方皆受益。
單次賽局的困境
但在單次賽局,合謀協議無法自我執行:
- 若對手維持 22.5,我偷偷增產到 Cournot 最適產量 $q_1^* = \frac{a-c}{2b} - \frac{22.5}{2} = 45 - 11.25 = 33.75$
- 偏離利潤 = $(100 - 10 - 33.75 - 22.5) \times 33.75 = 33.75 \times 33.75 = 1139.06 > 1012.5$
廠商有偏離誘因 → 合謀崩潰 → 回到 Cournot。
重複賽局的「以牙還牙」均衡
廠商反覆互動時,可用觸發策略(trigger strategy)維持合謀:
「只要對手合謀,我也合謀;一旦對手偏離,我永遠採 Cournot 產量報復。」
合謀可自我維持的條件(折現因子 $\delta$):
$$\underbrace{\frac{\pi^{cartel}}{1-\delta}}_{\text{合謀報酬}} \geq \underbrace{\pi^{deviate} + \frac{\delta \cdot \pi^{Cournot}}{1-\delta}}_{\text{偏離報酬}}$$
化簡得臨界折現因子:
$$\delta^* = \frac{\pi^{deviate} - \pi^{cartel}}{\pi^{deviate} - \pi^{Cournot}}$$
代入數值:
$$\delta^* = \frac{1139.06 - 1012.5}{1139.06 - 900} = \frac{126.56}{239.06} \approx 0.529$$
若廠商的實際折現因子 $\delta > 0.529$(即未來夠重要),合謀均衡可維持。
實務意涵:
- 市場成長快 → $\delta$ 高 → 合謀更穩定
- 市場透明度高(容易偵測偏離)→ 合謀更穩定
- 廠商數量多 → 每家偏離的個人利益更大,$\delta^*$ 升高 → 合謀更難維持
---
動能不穩定:Kinked Demand Curve
適用情境:解釋寡占市場價格僵固性(prices rarely change),但無法預測均衡價格水準。
核心邏輯:
- 若我漲價:對手不跟,我失去客戶 → 需求彈性高
- 若我降價:對手跟著降,我搶不到額外份額 → 需求彈性低
這造成需求曲線在現行價格處出現扭結(kink),對應的邊際收益曲線出現垂直缺口。
Price
| 需求曲線(現行價格以上,較平坦)
| /
P*|-----*
| \ (現行價格以下,較陡峭)
| \
| \
+-----------> Quantity
Q*
MR 曲線在 Q* 處有垂直缺口:
即使 MC 在缺口範圍內上移,最適產量不變 → 價格僵固限制:Kinked demand 只解釋現有均衡附近的僵固性,不解釋均衡如何形成,也未獲得強力實證支持。主要用途是描述性解釋,而非預測工具。
---
模型假設比較與選用陷阱
| 假設 | Cournot | Bertrand(同質) | Stackelberg |
|---|---|---|---|
| 決策變數 | 產量 | 價格 | 產量(序貫) |
| 決策時序 | 同時 | 同時 | 序貫 |
| 資訊 | 對手產量未知,但對稱信息 | 對手價格未知 | 追隨者觀察領導者 |
| 均衡概念 | Nash | Nash | Subgame Perfect Nash |
| 典型結果 | 介於競爭與壟斷 | 競爭結果($P=c$) | 先動優勢,比 Cournot 更多產出 |
常見錯誤:
- 看到「航空公司削價競爭」就用 Bertrand — 但航空公司有產能限制,應用容量限制 Bertrand或 Cournot(容量 ≈ 產量)
- 在動態市場用靜態模型 — Stackelberg 適合可觀察、可承諾的先動;若承諾不可信,Cournot 更適當
- 混淆「合謀均衡」與「Nash 均衡」— 合謀結果需要重複賽局或外部執行機制,單次賽局不成立
- 直接套公式而不確認邊際成本是否對稱 — 非對稱 MC 會改變均衡產量比例,領導廠商不一定是低成本廠商
Related skills
FAQ
What are the four market structures?
Perfect competition, monopolistic competition, oligopoly and monopoly, ordered by rising concentration and price power.
Does a unique product make a monopoly?
No; the market boundary decides. Smartphones are an oligopoly even if one firm's product is differentiated.