5D Library › Mechanism Design, Vickrey–Clarke–Groves Auctions and Incentive-Compatible Community Engineering · 中文

Chapter 7 Reading the Classics: One Reward, Settled Shares and “Rewards Reliable, Punishments Certain”

This chapter reads the earlier discussion of incentives, commitment and the assignment of rights beside a few texts from the Warring States to the Han: “one reward, one punishment, one teaching” in the “Rewards and Punishments” chapter of the Book of Lord Shang; “one hare runs, a hundred chase it” in its “Fixing Shares” chapter and in the Lüshi chunqiu, “Being Careful about Power”; and the sources of the phrase “rewards reliable, punishments certain” (xin shang bi fa) in the Han Feizi and the History of the Former Han. The aim is not to show that the ancients had mastered mechanism design, but to see how they expressed “rewards and punishments must be credible” and “rights must be settled first”, and to state plainly the basic differences of aim, premises and object between Legalist reward and punishment and the incentive compatibility of modern mechanism design. Book 27 already used “names and realities” from the “Two Handles” and “Fixing Laws” chapters of the Han Feizi; they are not repeated here.

Quotations follow Jiang Lihong’s Shangjunshu zhuizhi (蒋礼鸿 1986), Chen Qiyou’s Han Feizi xin jiaozhu and Wang Xianshen’s Han Feizi jijie (陈奇猷 2000; 王先慎 1998), Chen Qiyou’s Lüshi chunqiu xin jiaoshi (陈奇猷 2002) and the Zhonghua Shuju punctuated edition of the Han shu (班固 1962), cited by book and chapter only; the wording was also compared with common online transcriptions (the Chinese Text Project, the Project Gutenberg text of the Han Feizi, transcriptions on academic websites), with conventional punctuation. This round could not compare every character against the printed editions above; textual variants are noted in the text, and the author is asked to check against the printed editions. The English renderings consulted Pines’s translation of the Book of Lord Shang (Pines 2017).

The Book of Lord Shang, “Rewards and Punishments”: one reward, one punishment, one teaching

Source text

圣人之为国也,壹赏,壹刑,壹教。壹赏则兵无敌,壹刑则令行,壹教则下听上。夫明赏不费,明刑不戮,明教不变,而民知于民务,国无异俗。(蒋礼鸿 1986)

所谓壹赏者,利禄官爵抟出于兵,无有异施也。夫固知愚、贵贱、勇怯、贤不肖,皆尽其胸臆之知,竭其股肱之力,出死而为上用也。(蒋礼鸿 1986)

所谓壹刑者,刑无等级。自卿相将军以至大夫庶人,有不从王令、犯国禁、乱上制者,罪死不赦。有功于前,有败于后,不为损刑。……忠臣孝子有过,必以其数断。(蒋礼鸿 1986)

(Book of Lord Shang, “Rewards and Punishments”)

In plain language: the sage, in governing a state, unifies rewards, unifies punishments and unifies teaching. With unified rewards the army has no rival; with unified punishments orders are carried out; with unified teaching those below obey those above. Clear rewards need not be costly, clear punishments need not kill, clear teaching does not change, and the people know their tasks and the state has no divergent customs. “Unified rewards” means that profit, emoluments, office and rank come out of warfare alone, with no other dispensation. Then the clever and the stupid, the noble and the base, the brave and the cowardly, the worthy and the unworthy all exhaust the wisdom in their breasts and the strength of their limbs, and go to their deaths in the ruler’s service. “Unified punishments” means that punishment knows no ranks: from chief ministers and generals down to grandees and commoners, whoever disobeys the king’s orders, violates the state’s prohibitions or disturbs the ruler’s institutions is guilty of death without pardon. One who had merit before and fails afterwards does not have his punishment reduced for it. … Even a loyal minister or filial son who errs is judged by the measure of his offence.

Textual notes. In “抟出于兵”, 抟 is read as 专, “concentrated, single”. In the passage on “unified teaching”, some online transcriptions have 清瘘 where others read 清浊; this chapter does not quote that sentence. The author’s framework speaks of “‘rewards reliable, punishments certain’ in the ‘Rewards and Punishments’ chapter of the Book of Lord Shang”; on checking, what that chapter actually says is “壹赏,壹刑,壹教” and “明赏不费,明刑不戮”; the four characters 信赏必罚 are not in the Book of Lord Shang. Their sources are given below.

Reread in this book’s language, the passage has three layers, each only an analogy.

First, “profit and rank come out of warfare alone” is a one-dimensional incentive scheme: a single kind of action (military merit) earns rewards. Its advantage is a simple, verifiable rule (heads can be counted); its drawback is exactly the multitask problem discussed in Book 27: every activity that is not measured is abandoned, which the Book of Lord Shang does not hide, since it wants people to abandon activities such as “broad learning and eloquence”. Community points issued for a single countable action produce the same result (point farming in scenario E of Chapter 6).

Second, “past merit does not reduce punishment for a later failure” is a rule that merit and fault do not offset. In mechanism-design terms, rewards and punishments are two separate incentive constraints rather than one net score; this blocks the strategy of “first earn merit, then offend”.

Third, “clear rewards need not be costly, clear punishments need not kill” says that credible threats rarely need to be carried out in equilibrium: if everyone believes that disobedience means death, no one disobeys and the punishment need never be applied. This is structurally similar to the discussion of commitment in Chapter 1 and to Becker’s model in which deterrence depends on the probability of detection times the penalty (Becker 1968); part (B) of the program below gives the simplest calculation.

“Fixing Shares” and “Being Careful about Power”: one hare runs, a hundred chase it

Source text

一兔走,百人逐之,非以兔也。夫卖者满市,而盗不敢取,由名分已定也。故名分未定,尧、舜、禹、汤且皆如骛焉而逐之;名分已定,贪盗不取。(蒋礼鸿 1986)

(Book of Lord Shang, “Fixing Shares”)

慎子曰:「今一兔走,百人逐之,非一兔足为百人分也,由未定。由未定,尧且屈力,而况众人乎?积兔满市,行者不顾,非不欲兔也,分已定矣。分已定,人虽鄙,不争。」(陈奇猷 2002)

(Lüshi chunqiu, “Being Careful about Power”)

In plain language: when one hare runs, a hundred people chase it, not for the sake of the hare itself (but because whose it is has not been settled). Sellers fill the market, yet thieves dare not take their goods, because whose they are has been settled. So when shares are unsettled, even Yao, Shun, Yu and Tang would rush after it; when shares are settled, even greedy thieves do not take. (The words of Shen Dao quoted in the Lüshi chunqiu mean the same: hares are piled up in the market and passers-by do not even look, not because they do not want hares but because the shares are settled; once shares are settled, even a mean person does not contend.)

Textual notes. These lines of “Fixing Shares” have two readings. The received text in the Chinese Text Project and Project Gutenberg reads “非以兔也。夫卖者满市”, which the quotation above follows; an emended reading, “非以兔为可分以为百,由名(分)之未定也。夫卖兔者满市”, appears in some modern editions and website transcriptions and was probably supplied from quotations in encyclopaedic compilations. In “如骛焉而逐之”, some editions write 鹜 (duck) for 骛 (to gallop); it is generally read as 骛, and the book renders it “rush after”. Which reading Jiang Lihong’s Shangjunshu zhuizhi adopts in its text and notes could not be checked in this round and must be settled from the printed Zhuizhi. Shen Dao in “Being Careful about Power” reads “非一兔足为百人分也,由未定”, and later “积兔满市,行者不顾”. Only if “Fixing Shares” is read with the emended “卖兔者满市” (sellers of hares fill the market) is its difference from “积兔满市” (hares piled up in the market) worth comparing; with the received “卖者满市”, “Fixing Shares” is speaking of goods in general. The saying exists in several versions in pre-Qin texts and is generally attributed to Shen Dao; the date of the “Fixing Shares” chapter is also disputed, and this book takes no position.

The passage is directly analogous in structure to rent dissipation in modern economics: a valuable thing whose ownership is unsettled draws competitors to spend resources fighting over it, and the total cost of the fight can approach or equal its value. Tullock’s rent-seeking contest writes this as an equilibrium (Tullock 1980): nn people compete for a prize worth 1, each spending xix_i and winning with probability xi/∑jxjx_i/\sum_j x_j; in the symmetric equilibrium each spends (n−1)/n2(n-1)/n^2 and the total is (n−1)/n(n-1)/n. With a hundred people chasing one hare, 99 % of its value is consumed by the chase. “Settled shares” correspond to assigning property rights in advance: no one chases and nothing is dissipated.

The program below is the author’s model, illustrating this analogy and two simple calculations about reward and punishment: (A) Tullock’s contest for the hare; (B) the deterrence condition of “certain punishment”: stealing a hare worth 1 is deterred, with detection probability qq and penalty FF, if and only if qF≥1qF \ge 1; (C) “reliable rewards”: a promised reward for a task is paid only with probability ρ\rho, and people with costs uniform on [0,1] work only if ρR≥\rho R \ge cost, so getting 60 % to work requires an announced reward of 0.6/ρ0.6/\rho.

Calculation

# Chapter 7. The author's toy models beside "one hare runs, a hundred chase it"
# (Shangjunshu, Dingfen; Lushi chunqiu, Shenshi). An analogy, not a reconstruction.
# (A) Unassigned prize: n chasers spend effort x_i, win the hare (value 1) with
#     probability x_i / sum x (Tullock lottery). Symmetric equilibrium x* = (n-1)/n^2;
#     total effort (n-1)/n of the prize is dissipated. Checked by damped iteration of
#     the best reply sqrt(O) - O to the others' total O, then by a grid search for a
#     profitable deviation. Assigned prize ("names and shares settled"): no one chases.
# (B) Enforcement ("bi fa"): a thief takes a hare worth 1 unless deterred; expected
#     penalty q F (detection probability q, fine F). Deterred iff q F >= 1 (Becker 1968).
# (C) Credibility ("xin shang"): a reward R promised for a task, paid with probability
#     rho; agents with costs ~ U[0,1] work iff rho R >= cost. Share working = rho R.
def best_response(others, grid=4000):
    best = (0.0, 0.0)
    for k in range(grid + 1):
        x = 0.5 * k / grid
        u = (x / (x + others) if x + others > 0 else 0.0) - x
        if u > best[0]:
            best = (u, x)
    return best[1]

if __name__ == "__main__":
    print("(A) one hare, n chasers: equilibrium effort each, total effort (= share of")
    print("    the hare's value dissipated)")
    for n in (2, 3, 10, 100):
        x = 0.1
        for _ in range(5000):                       # damped best-reply iteration
            O = (n - 1) * x; x = 0.995 * x + 0.005 * max(0.0, O ** 0.5 - O)
        O = (n - 1) * x; dev = best_response(O)
        gain = (dev / (dev + O) - dev) - (x / (x + O) - x)
        print(f"    n = {n:3d}: iterated {x:.4f}, formula {(n - 1) / n ** 2:.4f};"
              f" total {n * x:.3f}; deviation gain {max(gain, 0.0):.4f}")
    print("    names and shares settled (owner known, theft punished): effort 0")
    print("(B) deterrence: q F >= 1 needed")
    for q, F in ((0.9, 1.5), (0.5, 1.5), (0.5, 3.0), (0.1, 3.0), (0.1, 12.0)):
        print(f"    detection {q:.1f}, penalty {F:5.1f}: expected penalty {q * F:5.2f}"
              f" -> {'deterred' if q * F >= 1 else 'not deterred'}")
    print("(C) to get 60 % of agents to work, announced reward needed = 0.6 / rho")
    for rho in (1.0, 0.8, 0.5, 0.25):
        print(f"    paid with probability {rho:.2f}: announce {0.6 / rho:.2f}")
# (A) one hare, n chasers: equilibrium effort each, total effort (= share of
#     the hare's value dissipated)
#     n =   2: iterated 0.2500, formula 0.2500; total 0.500; deviation gain 0.0000
#     n =   3: iterated 0.2222, formula 0.2222; total 0.667; deviation gain 0.0000
#     n =  10: iterated 0.0900, formula 0.0900; total 0.900; deviation gain 0.0000
#     n = 100: iterated 0.0099, formula 0.0099; total 0.990; deviation gain 0.0000
#     names and shares settled (owner known, theft punished): effort 0
# (B) deterrence: q F >= 1 needed
#     detection 0.9, penalty   1.5: expected penalty  1.35 -> deterred
#     detection 0.5, penalty   1.5: expected penalty  0.75 -> not deterred
#     detection 0.5, penalty   3.0: expected penalty  1.50 -> deterred
#     detection 0.1, penalty   3.0: expected penalty  0.30 -> not deterred
#     detection 0.1, penalty  12.0: expected penalty  1.20 -> deterred
# (C) to get 60 % of agents to work, announced reward needed = 0.6 / rho
#     paid with probability 1.00: announce 0.60
#     paid with probability 0.80: announce 0.75
#     paid with probability 0.50: announce 1.20
#     paid with probability 0.25: announce 2.40

In part (A) the iteration agrees with the formula and no profitable deviation is found on the grid: two chasers dissipate half the hare’s value, ten 90 %, a hundred 99 %. Part (B) shows that when detection is unlikely, deterrence can only be kept by heavier penalties (with detection 0.1, a penalty of 12 is needed), which is the logic of the Legalist case for “heavy punishment”; but Becker’s model equally warns that the heavier the penalty, the costlier a wrongful conviction, and this model ignores wrongful convictions. Part (C) shows that when the probability of paying a reward falls from 1 to 0.25, the announced reward must rise from 0.60 to 2.40 for the same effect: the cost of “rewards that are not reliable” is inflation of the promised rewards, the other face of the points inflation of Chapter 6.

The source of “rewards reliable, punishments certain”

Source text

晋文公问于狐偃曰:「寡人甘肥周于堂,卮酒豆肉集于宫,壶酒不清,生肉不布,杀一牛遍于国中,一岁之功尽以衣士卒,其足以战民乎?」……狐子对曰:「令无得不战。」公曰:「无得不战奈何?」狐子对曰:「信赏必罚,其足以战。」公曰:「刑罚之极安至?」……(陈奇猷 2000)

(Han Feizi, “Outer Collections of Sayings, Right, Upper”)

二曰必罚明威,三曰信赏尽能。(陈奇猷 2000)

(Han Feizi, “Inner Collections of Sayings, Upper: The Seven Techniques”)

孝宣之治,信赏必罚,综核名实,政事文学法理之士咸精其能。(班固 1962)

(History of the Former Han, eulogy to the “Annals of Emperor Xuan”)

In plain language: Duke Wen of Jin asked Hu Yan: “I share rich food with those in my hall and wine and meat with those in my palace; I give away the wine in the pots before it has even cleared and the raw meat before it has been spread out to dry; when I kill an ox the whole state shares it; a year’s harvest goes entirely to clothing the soldiers. Is this enough to make the people fight for me?” … Hu Yan answered: “Make it so they cannot but fight.” The duke asked: “How can I make them unable not to fight?” Hu Yan answered: “Make rewards reliable and punishments certain; that is enough to make them fight.” The duke asked: “How far must punishment go?” … Among the seven techniques by which the ruler controls his ministers in the “Inner Collections”, the second is “certain punishment to display authority” and the third “reliable reward to draw out ability”. Ban Gu judged the rule of Emperor Xuan: rewards were reliable and punishments certain, names were checked against realities, and so men of government, letters and law all became expert in their work.

Conclusion on the source. The earliest verifiable source of the four characters used together is Hu Yan’s answer to Duke Wen in the Han Feizi, “Outer Collections of Sayings, Right, Upper”; the “Inner Collections, Upper” lists “certain punishment” and “reliable reward” separately; and the eulogy of the “Annals of Emperor Xuan” uses the phrase to assess Emperor Xuan. The Book of Lord Shang speaks of “unified reward”, “clear reward” and “heavy punishment”, but not of these four characters. The Chinese subtitle therefore reads “the ‘Rewards and Punishments’ chapter of the Book of Lord Shang and the Han Feizi on rewards reliable, punishments certain”: the four characters come from the Han Feizi, while the “Rewards and Punishments” chapter supplies the structure of “unified reward, unified punishment, unified teaching”; the two are named side by side, and the phrase is not said to come from that chapter. (The parts of the “Outer Collections” passage replaced by ellipses are not quoted.)

Same name, different meaning: Legalist reward and punishment versus incentive compatibility

The word “incentive” links the two, but they answer different questions.

Aspect Legalist reward and punishment (Book of Lord Shang, Han Feizi) Incentive compatibility in mechanism design
Aim The ruler’s wealth, power and control: “with unified rewards the army has no rival” A social choice function declared in advance by the designer (efficiency, revenue or other), often judged by participants’ welfare
Participation No exit: the people cannot opt out of reward and punishment Usually requires individual rationality: taking part is no worse than not
What is elicited Conduct (farming, fighting, obeying), and whether ministers’ words match their deeds (forms and names) Private information (values, costs, preferences), on which the outcome is based
Who is bound Subjects; the ruler himself is not under the rewards and punishments (“punishment knows no ranks” lists ministers downwards) The designer: the revelation principle presupposes that the designer can commit to the rules
Instruments Rewards and punishments, with punishment weighing more Payments and allocations, usually without punitive personal consequences

The closest point of contact is credibility: “reliable reward” requires rewards to be kept and “certain punishment” requires punishments to be carried out, which is structurally similar to Chapter 1’s “the designer of a rule must be able to commit”; “clear rewards need not be costly, clear punishments need not kill” is similar to the logic of a deterrence equilibrium. But the Legalist “reliability” is a means of rule adopted by the ruler to make rewards and punishments effective, not a commitment to rights of the participants, and it is not constrained by their consent. Modern incentive compatibility requires that telling the truth or following the rules be the participant’s own best choice and (in most applications) that participation be voluntary. Calling Legalism “ancient mechanism design” erases this difference; this book says only that the two are structurally similar on two points: “rewards and punishments must be credible” and “rights must be settled first”.

Chapter summary

What the “Rewards and Punishments” chapter of the Book of Lord Shang says is “one reward, one punishment, one teaching”: rewards come only from military merit, punishment knows no ranks and merit does not offset faults, and “clear rewards need not be costly, clear punishments need not kill”. “One hare runs, a hundred chase it” in “Fixing Shares” and in Shen Dao’s words quoted in the Lüshi chunqiu differ in wording but agree in meaning: an unclaimed valuable thing provokes a contest, and the author’s model (Tullock’s contest) shows 99 % of its value dissipated when a hundred people compete, and nothing dissipated when ownership is settled. “Rewards reliable, punishments certain” does not occur in the Book of Lord Shang; its earliest source is Hu Yan’s words in the Han Feizi, “Outer Collections of Sayings, Right, Upper”, and it recurs in the eulogy of the “Annals of Emperor Xuan”. Legalist reward and punishment serves the ruler, allows no exit and binds the subjects rather than the ruler; the incentive compatibility of mechanism design rests on the participant’s own best choice and voluntary participation. The two share a name but not a meaning, and the similarity is only structural.

References

Becker, Gary S. 1968. “Crime and Punishment: An Economic Approach.” Journal of Political Economy 76 (2): 169–217.
Pines, Yuri. 2017. The Book of Lord Shang: Apologetics of State Power in Early China. Columbia University Press.
Tullock, Gordon. 1980. “Efficient Rent Seeking.” In Toward a Theory of the Rent-Seeking Society, edited by James M. Buchanan, Robert D. Tollison, and Gordon Tullock. Texas A&M University Press.
王先慎. 1998. 韩非子集解. 钟哲点校. 新编诸子集成. 中华书局.
班固. 1962. 汉书. 中华书局.
蒋礼鸿. 1986. 商君书锥指. 新编诸子集成. 中华书局.
陈奇猷. 2000. 韩非子新校注. 上海古籍出版社.
陈奇猷. 2002. 吕氏春秋新校释. 上海古籍出版社.

Fan, Y. (2026). Mechanism Design, Vickrey–Clarke–Groves Auctions and Incentive-Compatible Community Engineering: From Myerson's Optimal Auction and the Hurwicz Impossibility Theorem to Sybil-Resistant Reputation Games, Read Beside Classical Legalist Treatises on Rewards and Incentives [机制设计、维克里—克拉克—格罗夫斯(VCG)拍卖与数字社群激励相容工程] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23836274 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).