← How an Economy Grows and Why It Crashes
第 1 章 · 20 分钟
The minimal model: output, postponed consumption, tools
Three links, and the fewest assumptions that make them hold.
概念地图
Roundabout
Build tools first, then final goods; the cost of the tools is consumption postponed.
深色为本章概念，浅色为其他章
陪练模式
不想只是读？让它带你走一遍。
导师把这一章拆成小步，每一步先问你，再讲；你用自己的话答，它先认下对的部分再纠偏。用你自己的 AI，对话只存在这台设备上。
Locus01
延伸阅读
Böhm-Bawerk · 1880s · The Positive Theory of Capital
Wrote 'production takes time' into value theory: the longer route through tools yields more final output from the same labour. He argued against theories that made interest a form of extraction; the cost is that the whole case rests on 'more roundabout is generally more productive', a premise never independently tested.
Foundational
Wicksell · turn of the century · natural rate versus market rate
Connected Böhm-Bawerk's real picture to money: a bank rate away from the natural rate keeps pushing prices. This gave capital theory an interface that monetary policy can disturb, and made it hard ever after to test the theory apart from monetary phenomena.
Turn
Hayek · 1931 · Prices and Production
Arranged capital into stages by distance from final consumption, with the interest rate coordinating the proportions. The Keynesian reply went straight at the core: the stages cannot be uniquely identified in the statistics.
Turn
Schiff · 2010 · How an Economy Grows and Why It Crashes
Rewrites that line as a fable for non-specialists. The fable buys clarity by dropping the disputes — and so the three failure points above never appear in the book.
Subject of this course
机制02
The postponement pump
On the island each person catches one fish a day by hand and eats it. To build a net somebody has to eat less on some day, and the fish not eaten is saving. The book makes this its first brick: not a number in an account but a postponement, and that postponement is exactly the food that feeds the net-building days. Whoever bore the postponement has a claim on the extra fish the net later brings in — the same thread Böhm-Bawerk used to explain interest.
机制Postponed consumption, through the toolmaking time it feeds, raises future capacity.
可迁移性测试
Move it to a software team: the outsourcing work not taken this quarter is the uneaten fish; the freed engineer-days pay for the refactor, after which the same headcount ships more. The isomorphism breaks in one place — a code 'net' rots on its own as dependencies move, a fishing net does not.
机制03
Length of the detour
Whether the builder skips fishing for one day, two days or thirty decides whether he ends up with a hand net or a trawl. Böhm-Bawerk called that length the degree of roundabout production. What the book leaves unsaid is that the builder does not set this length on his own — its ceiling is how much stored fish everyone else is willing to hand over. Chapter 2 is entirely about what happens when that ceiling is crossed.
机制The available stock, through the waiting time it can cover, bounds how elaborate a tool can be.
可迁移性测试
Move it to clinical research: the length of a phase-three trial is set not by the investigator's ambition but by how many years of running costs the funding covers. The isomorphism breaks in that a trial stopped midway usually yields nothing, while half a net still catches something.
Derivation04
From two assumptions to 'growth must be preceded by less consumption'
What is being derived is the full form of the book's opening claim: before output rises, consumption must fall for a stretch.
Labour force fixed; technology unchanged until the tool is built
Chosen. Relax it — let technology be imported — and the conclusion changes shape at once; chapter 4 meets this.
Tools can be built only out of output not consumed this period
Chosen, and the heaviest of the three. It rules out building first and saving afterwards, which is exactly where chapter 2 parts company with the mainstream.
(read: output in period ) does not rise from the act of building
Chosen: it assumes the labour pulled into building was not idle before. Chapter 3 attacks this one on its own.
推导 · 0 / 4
裂缝05
Cracks in this framework
争议地形06
本书主张
Saving is where growth starts: without postponed consumption there are no tools and no higher output. Stimulating consumption changes where output goes, not how much there is.
另一种看法
The line from Keynes's General Theory of 1936 holds that when capacity is idle, spending determines output rather than the reverse; more saving then lowers total income and realised saving can fall — the paradox of thrift.
分歧扎在
The disagreement is rooted in one assumption: whether capacity is fully used. The book assumes everyone on the island fishes every day; the Keynesian side assumes involuntary unemployment. Each side reasons correctly from its own premise.
什么证据能裁决
What would settle it is direct measurement of the output gap: for comparable shocks, compare the real output path 6–12 quarters after a rise in the saving rate in high-slack and low-slack economies. If the paths do not differ systematically, the capacity-utilisation split itself does not carry the argument.
The evidence points to 'it depends': near full capacity the book's mechanism is closer to the facts, in a deep contraction the Keynesian one is. What is missing is a measure of capacity utilisation both sides accept in advance — the existing measures often disagree on the sign of the gap at the same date.
Falsification07
The postponement pump as a preregistrable test
The mechanism under test is the postponement pump: postponed consumption, through the toolmaking time it feeds, raises future capacity.
Data source: gross saving rate from the World Bank WDI; capital stock and labour productivity from the Penn World Table
Public sources with published definitions, so the test can be rerun unchanged by someone else.
Sample: economies once they pass into upper-middle income, full sample, no ex post exclusions
Dropping 'special cases' is the most common entry point for a degrees-of-freedom pump; it is sealed off in advance.
Window: labour productivity in years 5 to 10 after a change in the saving rate
Long enough for tools to be finished, short enough not to be swamped by the next shock. The bounds are set in advance, not tuned.
Threshold: the saving-rate coefficient is significantly positive after controlling for initial productivity
Registered in advance; no redefinition after the result is seen.
Failure condition: the coefficient is insignificant, wrong-signed, or holds only once some group of countries is dropped
The third case counts as failure — it means the result rests on the exclusion rule, not on the mechanism.
推导 · 0 / 3
接口08
Which slot it hangs on
挂在哪个槽位The conclusion you already hold is probably 'save now so you can spend more later' — a claim about personal timing. This chapter does not touch it; it touches the slot next to it: capacity.
Does this chapter (a) replace your existing timing rule about saving and spending, or (b) open a new slot beside it for 'where physical capacity comes from'?
慢变量Register two slow variables: capital spending as a share of revenue in your industry, and output per employee. In a few days look at the first one and whether it has risen or fallen over three years.
小结09
本章小结
01The first two links of saving–tools–output are identities; only the third carries empirical risk.
02The chain needs three assumptions: fixed labour, tools built only from unconsumed output, output unchanged while building.
03The third fails when capacity is idle, which is where the Keynesian counterexample lands.
04The framework has no ex ante criterion for whether a tool truly raised capacity; that piece is unfalsifiable.
05A preregistrable test settles correlation and magnitude, not causal direction.
提取练习 · 合上书，先自己答一遍。
?Is 'output equals consumption plus saving' (a) an identity or (b) an empirical claim?
?Does 'imported equipment raises local capacity' weaken (a) the first assumption or (b) the second?
Forced choice10
A · Identify the mechanism
Which of these is a mechanism?
二选一
Forced choice11
C · Locate the crack
In which situation does this framework give a confident and wrong answer?
二选一
读到这里，把它记为已读。