前回エントリで紹介したワルドマンのベンチマークモデル論と通底していなくもない議論をEveryday Economistブログが展開している(H/T Economist's View)。
ブログ主(Josh Hendrickson)は、RBCモデルの悪名高い(?)生産性の変動要因について、以下のような解釈を示している。

The idea behind the real business cycle model is that fluctuations in aggregate productivity are the cause of economic fluctuations. If all firms are identical, then any decline in aggregate productivity must be a decline in the productivity of all the individual firms. But why would firms become less productive? To me, this seems to be the wrong way to interpret the model. My preferred interpretation is as follows. Suppose that you have a bunch of different firms producing different goods and these firms have different levels of productivity. In this case, an aggregate productivity shock is simply the reallocation from high productivity firms to low productivity firms or vice versa. As long as we think of all markets as being competitive, then the RBC model is just a reduced form version of what I’ve just described. In other words, the RBC model essentially suggests that fluctuations in the economy are driven by the reallocation of inputs between firms with different levels of productivity, but since markets are efficient we don’t need to get into the weeds of this reallocation in the model and can simply focus our attention on a representative firm and aggregate productivity.
リアルビジネスサイクルモデルの背後にある考え方は、マクロ的な生産性の変動が経済の変動を引き起こす、というものである。もしすべての企業が同一ならば、マクロ的な生産性の低下は、個々の企業すべての生産性の低下ということになる。しかしなぜ企業の生産性が低下するのか? 私に言わせれば、それはモデルの解釈の仕方が間違っている。私の解釈は次のようなものである。たくさんの様々な企業が様々な商品を生産していて、各企業の生産性の水準は異なっているものとしよう。この場合、マクロ的な生産性ショックは、生産性の高い企業から低い企業への再配分、もしくはその逆に過ぎない。すべての市場が競争的であると考える限り、RBCモデルは以上の話の単なる誘導形ということになる。言い換えれば、RBCモデルは基本的に、経済の変動は生産性水準が様々な企業の投入の再配分によって引き起こされているが、市場が効率的であるために我々はこの再配分の詳細にモデルで立ち入る必要は無く、代表的企業とマクロ的生産性だけに関心を集中することができる、ということを述べているのである。


I think that my interpretation is important for a couple of reasons. First, it suggests that while “forgetting how to use technology” might get chuckles in the seminar room, it is not particularly useful for thinking about productivity shocks. Second, and more importantly, this interpretation allows for further analysis. For example, how often do we see such reallocation between high productivity firms and low productivity firms? How well do such reallocations line up with business cycles in the data? What are the sources of reallocation? For example, if the reallocation is due to changes in demographics and/or preferences, then these reallocations could be interpreted as efficient responses to structural changes in the economy and be seen as efficient. However, if these reallocations are caused by changes in relative prices due to, say, monetary policy, then the welfare and policy implications are much different.
Thus, to me, rather than denigrate RBC theory, what we should do is try to disaggregate productivity, determine what causes reallocation, and try to assess whether this is an efficient reallocation or should really be considered misallocation. The good news is that economists are already doing this (here and here, for example). Unfortunately, you hear more sneering and name-calling in popular discussions than you do about this interesting and important work.
私の解釈は幾つかの点で重要だと思われる。第一に、「技術の使い方を忘れた」という解釈は講義中に笑いを取れるかもしれないが、生産性ショックを考える際にはあまり有用ではない、ということが言える。第二に、より重要な点として、この解釈はさらなる分析を可能にする。例えば、そうした高生産性企業と低生産性企業の間の再配分を我々はどの程度頻繁に目にするだろうか? そうした再配分は景気循環のデータとどの程度うまくマッチするだろうか? 再配分の原因は何だろうか? 例えば、もし再配分の原因が人口動態ないし嗜好の変化にあるならば、そうした再配分は経済の構造変化への効率的な反応と解釈することができ、効率的なものである、と捉えられる。一方、もしそれらの再配分が、例えば金融政策のせいで相対価格が変化したために引き起こされたならば、厚生や政策にとっての意味合いがかなり変わってくる。

これはまさに、RBCモデルをワルドマンの言う「ベンチマークモデル(benchmark model)」として使い、生産性の異なる企業間の再配分を導入したモデルを「現実に近付けたモデル(models which we think might be useful approximations to the truth)」として両者を対照させる、という考え方になるかと思われる。続く最終段落ではその点が明確に述べられている。

Finally, I should note that I think one of the reasons that the real business cycle model has been such a point of controversy is that it implies that recessions are efficient responses to fluctuations in productivity and counter-cyclical policy is unnecessary. This notion violates the prior beliefs of a great number of economists. As a result, I think that many of these economists are therefore willing to dismiss RBC out of hand. Nonetheless, while I myself am not inclined to think that recessions are simply efficient responses to taste and technology changes, I do think that this starting point is useful as a thought exercise. Using an RBC model as a starting point to thinking about recessions forces one to think about the potential sources of inefficiencies, how to test the magnitude of such effects, and the appropriate policy response. The better we are able to disaggregate fluctuations in productivity, the more we should be able to learn about fluctuations in aggregate productivity and the more we might be able to learn about the driving forces of recessions.

*1:cf. 本ブログの関連エントリ