というNBER論文が上がっているungated版へのリンク)。原題は「An 'Austrian' Model of Global Value Chains」で、著者はPol Antràs(ハーバード大)。27日エントリで紹介した同じ著者の論文と似たようなタイトルだが、ここではそちらで考慮していなかったサプライチェーンに焦点を当てたとの由。

I develop a stylized model of multi-stage production in which the time length of each stage is endogenously determined. Letting the production process mature for a longer period of time increases labor productivity, but it comes at the cost of higher working capital needs for firms. Under autarky, countries with lower interest rates feature longer production processes, higher labor productivity, and higher wages. In a free trade equilibrium, countries with lower interest rates specialize in relatively ‘time intensive’ stages in global value chains (GVCs). Yet, if free trade brings about interest rate equalization, wages are also equalized and the pattern of trade is instead shaped by capital intensity and capital abundance, regardless of the time intensity of the various stages. Reductions in trade costs lead to patterns of specialization associated with higher amounts of vertical specialization in world trade. A worldwide decline in interest rates similarly fosters an increase in the share of GVC trade in world trade. The framework also sheds light on the role of trade credit and trade finance in shaping international specialization.


「Instantaneous Inflation」という最近の論文の分析が広く引用されている、とクルーグマンツイートしているクルーグマンが引リツしているJan Eeckhoutはポンペウ・ファブラ大学の教授で、分析内容を連ツイで解説するとともに、自サイトに上げている論文にリンクしている。




I've been a bit behind on commenting on this widely cited recent analysis, but thought I might still have something to contribute. Warning: very wonkish, not for normal humans 1/
First off, I basically agree with Eeckhout's conclusion: underlying inflation has fallen fast, although not quite ready to say that it's 2 percent. Also agree with the general point about the need to smooth but not too much: monthly is too noisy, annual too far behind reality 2/
In fact, most practical analysts are using a variety of methods to try and extract the inflation signal from the noise, typically looking at 3-month changes in an index that excludes volatile stuff like energy and stuff with known long lags, mainly shelter 3/
What does Eeckhout's procedure add? It doesn't throw out data before past 3 months. Maybe more important, "artisanal" inflation measures can tempt analysts into choosing measures that tell them what they want to hear. So there's something to be said for a hands-free filter 4/
On the other hand, the filter also throws out information. The dynamics of, say, official food and shelter price indexes are different in known ways; applying a common smoothing approach to the overall index means ignoring what we know about sectors 5/
A parallel that immediately occurred to me is estimating potential GDP. It can seem appealing to use a statistical smoothing procedure like Hedrick-Prescott, which sometimes produces plausible results. But we also know that it can go badly wrong at times 6/
E.g., a smoothing approach interprets any sustained downturn as permanent — so it ends up saying that the Great Depression was over by 1935. I actually worry Eeckhout's approach may be doing a milder version of this now, over-interpreting favorable recent inflation shocks 7/
I say this even though I want to believe that inflation has rolled over. It's clearly way down, but a mechanical approach throws out data that might give reason for caution, say on wages 8/
All that said, very happy to see this kind of work being done. Traditional core inflation has performed badly lately, and we need to be exploring new approaches 9/
Eeckhoutの手法の付加価値は何か? その手法では3か月より前のデータも捨て去ることはしない。それよりも重要かもしれないのは、「職人芸による」インフレ指標では、耳にしたいことを伝える指標を選択する誘惑に分析者が駆られる可能性があることだ。ということで、人の手を介さないフィルタには何らかの価値がある。


*2:cf. ここここここ


「Secular stagnation is not over」という小論をブランシャールが書きクルーグマンが賛意を表している






*1:この点についてはジョン・コクランAndy Harlessが違和感を感じた旨のツイートをしている。そのうちのHarlessは、長期停滞で問題になるのはrがgに比べて低いことではなくr*に比べて高いことではないか、と指摘した。それに対しブランシャールは、近著を読んでほしいと応じつつ、低いr*にはr*<gという側面と、ゼロ金利下限のために中銀がrを十分に下げられないためにr*<rとなるという2つの側面がある、という見解を示している

*2:上述のブランシャールの以前のブログエントリでは金融財政政策についての彼の考えが45項目にまとめられているが、その第44項では日本を取り上げ、政府が30年に亘って巨額の財政赤字を計上して債務GDP比率が上昇した一方で、日銀が政策金利を実効下限に留めたことは、意図したものではなかったかもしれないが、条件付きで正しい戦略だったと言える、と評価している。その前後の項では、財政出動が過小だったようにみえる金融危機後の欧州(A case of too little?)と、過大だったようにみえるコロナ禍時の米国(A case of too much?)を取り上げており、それらとの対比で日本は適正だったのでは(A case of just right?)、と述べている。ただ、日本の債務の持続可能性には彼も懸念を示し、需要を引き上げる別の方法を探すことが急務、とも述べている。



@paulkrugman is right that measuring inflation is difficult. But he continues his pattern of choosing to emphasize issues that could lower rather than raise expected inflation. He ignored the housing issues when I and others were pointing them out last year.
The same point about new transactions vs. past transactions present wrt housing applies to the labor market but Paul fails to note that new hire wages are still way up and reservation wages continue to climb. He doesn’t consider that effort may be down for those working at home.
Nor does Paul note that some of the deflation seen in used cars and other sectors is not sustainable because it reflects the unwinding of past price run ups.
There is the further point that non availability is much more common than several years ago and implies that “real prices” are higher than measured ones.
It may well be that given the balance of risks the @federalreserve should stop hiking soon but the debate is best framed recognizing all the biases in official measures.


これにクルーグマンは、Joey Politanoという経済ライターのこちらのツイートをRTすることで応じている。

Incredibly ironic that while chiding Paul Krugman for overlooking data lags in housing CPI last year Larry Summers cites data that is a 12-month moving average of observed wage growth rates. In other words, more lagging data.


Some notes on inflation. First, clearing up a misconception: no, measures like "supercore" that exclude food, energy, shelter and used cars aren't finding disinflation by throwing out the inflation 1/
Headline inflation is actually being held down by some clearly one-off things, like falling gas prices, but also being exaggerated by shelter prices that reflect a rent surge that ended many months ago. Overall it's more or less a wash, and the disinflation is real 2/
Second, the current inflation debate is very different from the one that took place in 2021. Then, Team Transitory argued that despite large fiscal stimulus we wouldn't see a big rise in inflation; we were wrong, and have admitted that 3/
But now the argument is between what we might call Team Stagflation and Team Soft Landing. Many of the same players, but in many ways a role reversal in arguments 4/
At this point the big surge in domestic spending is behind us, with nominal spending growth rapidly slowing to rates more or less consistent with target inflation 5/
Team Stagflation nonetheless argues that inflation will remain high, that bringing it down will require an extended period of high unemployment, as happened in the 1980s. And of course they could be right. But what's their theory of the case? 6/
Textbook macro says that inflation can become entrenched via expectations. But medium- and long-term inflation expectations never rose much, and are now back to roughly pre pandemic levels 7/
You could argue that the textbook model is all wrong, and inflation will be sticky for other reasons. But most of the data seem to show inflation falling fast, and attempts to deny the good news are getting increasingly contorted 8/
We are still waiting for Godot, I mean the Employment Cost Index, which might (or might not) give some new justification for pessimism. But as Joey Politano says, the available data suggest deceleration 9/

My hope is that people who were right about inflation — and got a lot of justified credit for making that call — won't try to hold on to their glory days by denying the growing evidence that the inflation surge is behind us 10/



(Joey Politanoのツイート*2


*1:cf. ここ



というECB論文をMostly Economicsが紹介している。原題は「Using machine learning to measure financial risk in China」で、著者はAlexander Al-Haschimi(ECB)、Apostolos Apostolou(IMF)、Andres Azqueta-Gavaldon(Sensyne Health*1)、Martino Ricci(ECB)。

We develop a measure of overall financial risk in China by applying machine learning techniques to textual data. A pre-defined set of relevant newspaper articles is first selected using a specific constellation of risk-related keywords. Then, we employ topical modelling based on an unsupervised machine learning algorithm to decompose financial risk into its thematic drivers. The resulting aggregated indicator can identify major episodes of overall heightened financial risks in China, which cannot be consistently captured using financial data. Finally, a structural VAR framework is employed to show that shocks to the financial risk measure have a significant impact on macroeconomic and financial variables in China and abroad.

*1:cf. 日本語による紹介記事

*2:cf. Topic model - Wikipedia



Mostly Economics経由のオーストリア学派関連論文をもう二丁。一つは、アリゾナ州立大のScott Scheallによる「Complexity, Policymaking, and The Austrian Denial of Macroeconomics」。以下はその要旨。

Economists associated with the Austrian School of Economics are known to deny the value of macroeconomics as descended from the work of John Maynard Keynes and, especially, his followers. Yet, Austrian economists regularly engage in a related scientific activity: theorizing about the causes and consequences of economic fluctuations, i.e., the business cycle. What explains the Austrians’ willingness to engage in theorizing about the business cycle while denying the scientific import of macroeconomics? The present paper argues that the methodological precepts of the School, which have remained largely in place since Carl Menger first pronounced them at the start of the Methodenstreit, justify the kind of business-cycle theorizing that Austrians do and imply the limited scientific value of macroeconomics as descended from Keynes.
経済学のオーストリア学派に属する経済学者は、ジョン・メイナード・ケインズの研究ならびにとりわけ彼の信奉者の研究に端を発したマクロ経済学の価値を否定することで知られている。しかしながら、オーストリア学派の経済学者は、それと関連する科学活動を常に行っている。経済変動、即ち景気循環の原因と結果の理論化である。マクロ経済学の科学的重要性を否定する一方で、景気循環について理論化しようとするオーストリア学派の意図はどのように説明されるのか? 本稿は、同学派の方法論上の教えについて論じる。その教えは、方法論争*1の開始時にカール・メンガーが最初に明確化して以来、概ねそのままの形を保っており、オーストリア学派が行っているようなタイプの景気循環の理論化を正当化し、ケインズに端を発するマクロ経済学の科学的価値が限定的であることを含意する。


As an attempt to come to grips with some massively complex phenomena, ABCT explains the principle that gives rise to, and predicts patterns to be observed in, business-cycle phenomena, but it does not aspire to a full explanation or a precise prediction of particular instances of the trade cycle. Indeed, an implication of the theory and the methodological principles upon which it is built is that such explanations and predictions are beyond the ken of cognitively limited human beings. Built upon the methodological principles of theoreticism, individualism, and subjectivism, as well as a profound appreciation of the limitations that complexity places on the possibilities for explaining and predicting, and, therefore, controlling via political means, business-cycle phenomena, there is little prospect that Austrian Business Cycle Theory could ever serve as a tool of political mischief, unlike the oversimplified methods of macroeconomics that encourage economic policymakers to pretend to knowledge they do not possess.

もう一つは、ミドル・テネシー州立大のDaniel J. Smithによる「Austrian Economics as a relevant research program」。以下はその要旨。

What is the relevancy of modern Austrian economics? Austrian economics, from its origins, has attempted to push economics towards greater relevance by developing and refining a methodological approach that enhances the operational validity of its scientific conclusions for decision-making in the real world. In a theoretical paradigm, this led to the development of theoretical insights on significant economic phenomena often excluded from static economics models. As the economics profession took an empirical turn, modern Austrian economics has demonstrated its continued relevancy through empirical methods that apply economic theory to pressing, long-standing policy issues.
現代オーストリア学派経済学の意義は何だろうか? オーストリア学派経済学は、その最初から、現実世界における意思決定についての科学的結論の実践的有効性を増すような方法論的アプローチを発展させ洗練させることで経済学の重要性を高めようとしてきた。理論パラダイムにおいてそのことは、静学的な経済モデルからはしばしば除外されてきた重要な経済現象についての理論的洞察の発展につながった。経済学者が実証志向に転じるにつれて、現代オーストリア学派経済学は、差し迫った長期に亘る政策問題に適用される実証的手法を通じて、今も続くその重要性を示してきた。

Mostly Economicsは論文から以下の一節を引用している。

…what I mean by relevance is specifically that we do scientific research with policy relevance and that we arrive at scientific conclusions that do not rely on limiting assumptions to the extent that would lead to potentially adverse unintended consequences if used today or in the future, as a basis for real-world decision-making (see, for instance, Munger 2022). It is the distinction, according to Radomysler (1946, p. 198), “between theory that does apply to reality and theory that does not.” An academic can eschew public policy and be wholly concerned only with long-standing problems in political economy but still seek to produce relevant research under this definition.


というNBER論文が上がっているungated版へのリンク;H/T Mostly Economics)。原題は「Interest Rates and World Trade: An 'Austrian' Perspective」で、著者はPol Antràs(ハーバード大)。

This paper develops a framework to study the interplay between world trade and interest rates. The model incorporates an explicit notion of time and of production length, along the lines of the ‘Austrian’ tradition of Böhm-Bawerk (1889). Changes in the interest rate affect production lengths, labor productivity, and the financial costs of exporting. I decompose the response of the volume of world trade to changes in the interest rate into four components: (i) a labor productivity effect, (ii) a propensity to consume out of labor income effect, (iii) a temporal dimension of variable trade costs effect, and (iv) a selection into exporting effect.

ungated版には「Böhm-Bawerk Meets Krugman」という章があり、ベーム=バヴェルクのほかにクルーグマンの貿易理論を取り込んだことが記されている。


  • 生産過程の熟成は労働生産性を高めるが、その対価として企業が必要とする実働資本が増える
  • 外市場での販売は企業所得の源泉を増やすが、輸出においては生産と消費の間のタイムラグが増える
  • 金利の変化は生産期間、労働生産性、輸出の財務費用に影響する


  1. 生産期間と労働生産性を高め、それによって賃金率も高める
  2. 労働所得からの消費性向を高める
  3. 輸出の変動費に関連する実働資本の追加必要分を減らす
  4. 輸出の固定費に関連する実働資本の追加必要分を減らす*1