A new book by Robert Allen must be devoured in one or two sittings, like a detective novel. Then it is ready for rereading because the same themes and questions keep cropping up in so many facets of the process of growth. Sure enough, The British Industrial Revolution in Historical Perspective is important, satisfying, and ultimately disquieting.

The book is important because it follows a line of inquiry with a relentless focus that reminds us that the “economic” in economic historian is of considerable importance. Allen wants to convince us that technological change is a production process where increased inputs (labor and capital) yield more output (new usable technologies). Ideas are well and fine, but the businessmen of the eighteenth century wanted techniques for their businesses. If the Industrial Revolution was largely the elaboration of techniques, that used more capital and more energy and less labor and land, then it ought to have happened when and where interest rates and energy price were low relative to wages and land rents high. That relative prices were key to the Industrial Revolution is a simple idea, but it is a very powerful one. This thesis is not particularly novel, but Allen's purpose is deploy a database of prices to evaluate its quantitative merits.

Hence the book has two parts. The first devoted to getting the reader to his rendezvous with destiny: the 1700s when Darby wrestles with coke, Akrwright with mules, and Bolton with pistons. In the second part, Allen traces the interconnection between prices and the development of new technologies.

In chapter 2, we learn that England alone escaped the wage stagnation or decline that beset Europe in the century that followed the Black Death. After that cataclysm, wages were high everywhere, but they began an inexorable decline almost everywhere. We then move on to reverse the old Marxian argument of an industrial revolution preceded by an agricultural revolution (chapter 3). Allen in this way makes his work harder, indeed he is forced to leave aside some important conclusions of Enclosure and Yeoman, namely that although enclosures did not spur productivity growth, the seventeenth century was a period of rapid gain in agricultural efficiency. In this volume, wages are explained by urbanization and the capture of commercial routes, agriculture is just another sector that responds to changes in demand both by changing output and changing techniques. England did not just benefit from high wages and cheap food, but as we learn in chapter 4, it also was able to maintain the cost of energy at a low level by switching to coal both earlier and more extensively than anywhere else than in Europe. Doing so required a variety of adaptations, in particular in the domain of home heating (which as Wrigley reminded us long ago was the largest use for coal). Chapter 5 then proceeds to estimate a system of equations that predict wages, agricultural productivity, urbanization, and proto-industrialization. The statistical effort leads Allen to conclude that what drove London wages up was international trade, while most other covariates were unimportant. In particular, the English commercial fleet's miraculous appropriation of what had been Dutch business is critical to explain why London pulls ahead of Amsterdam. This reader did not find this chapter extremely satisfying because neither the temporal (once a half century) nor spatial dimensions of the observation was justified. Can we have an observation for Italy if we have one for Belgium and one for the Netherlands?

In the second part, Allen traces out the consequence of these prices for the development of new technologies. Chapter 6 lays out the induced innovation approach in clear and concise terms, and generously connects it with debates in the history of technical change. Chapter 7 takes on the steam engine and argues that only where energy was extraordinarily cheap would it pay to install the earliest and least efficient machines. Chapter 8 tackles cotton and explicitly investigate the question of whether it would have paid to build spinning jennys in France. The answer is no: labor there was too cheap to warrant the capital labor substitution implied by the new devices. The demonstration is repeated with Coke in chapter 9. Chapter 10 then takes on Joel Mokyr's recent narratives of technical change that emphasize the importance of that French idea: enlightenment. To do so, Allen borrows from Zorina Khan and Ken Sokoloff and culls the biographies of Great British inventors to demonstrate that most of them were moneymaking rubes who might have acquired and enlightenment veneer after being successful.

All this is presented in 275 pages of lively and extremely satisfying prose. The questions are defined precisely, the path Allen takes to answer them transparent, and the answers neatly laid out along with their implications of the larger argument. This is the book you should use to teach the Industrial Revolution, doing so will allow you to add half a dozen articles of your choosing to push the student into other dimensions of the problem of growth.

But you should be worried. To begin, the volume relies exclusively on models that assume (but never discuss why) economies are closed. Trade is simply a high-income sector; the fact, for example, that cotton is imported plays no role in Allen's story. Consider the elegant but poorly paid French spinners at their wheels, the workaday English Jenny operators, all produce similar yarn. In a world where markets are integrated (and these should have been because the cotton came from the Americas, and the cost of getting the finished product to the consumer small), only one of these technologies should have prevailed. One of three things should have happened. (1) As wages rose in England, Lancashire manufacturer should have outsourced their spinning to Normandy, or genial English entrepreneurs should have been installed their jennies in Normand rather than England (combining the low cost of English capital with low French wages), or the English jennies should have sent all French spinning wheels to the scrap heap. Somehow in a world without institutions (no tariff policy, no market imperfections), the third outcome seems unlikely and the one that occurred inconceivable. If you doubt my conclusion, consider contemporary outsourcing decisions and the decline of manufacturing in the West.

The thorny issue of trade then leads us to ponder why England, France, Italy, or Spain are relevant economic concepts or units over the long run. At some point in time, some of these names refer to places which had common economic policies (like tax rates or tariff policies) and common institutions (laws, regulations, and so forth), or even common foreign policy. But consider the United Kingdom. From 1600 to 1850 England and Ireland and Scotland had common foreign policy, but it was not until the acts of union of 1707 and 1801 that they shared common tariff policies and increasingly common regulatory frameworks. When we say wages are high in England in 1750, are we thinking of a region in some larger political entity whose definition comes from some economic logic? For instance, England in the United Kingdom stands as the hinterland for London. Alternatively, are we thinking of England as policy unit recently merged in the United Kingdom? When we take wages for England to be those of London, what should we do about France? Does Paris stand for France or something called Ile de France, or the fiscally integrated Cinq Grosses Fermes. Should Florence stand for the yet to be born Italy, or the Grand Duchy of Tuscany? The economic implications of our historical assumptions are far from trivial. If London's extremely high wages were beneficial to England, was it because London could capture a large share of European commerce and the size of its hinterland is irrelevant, or was it because institutions insured that the Scotts and the Irish would not get the same benefits as the English? Comparative economic history, like Allen's, takes geography as an important variable but relying on the convenience of current sovereign state makes it impossible to examine a whole host of issue about the interaction between urban growth and regional economic development. In other words, Allen's work should encourage us to remember that the definition of the spatial units of our analyses is critical to how our research proceeds.

If we accept this version of history, England and English economic history is utterly irrelevant to the problem of global economic change. At least as this reader sees it, since 1800 the dominant public policy problem has been to spur economic growth and the adoption of productivity enhancing technique. In contrast, the magical alignment of relative prices in England in 1700 is a historical accident that is interesting only because of our curiosity for our origins. After 1850 other countries (as Allen points out) were able to industrialize without having to first achieve high wages or low capital costs. Thanks to English technology, everywhere else relative price change followed technical change.

Moreover, at least in Allen's view, public policy in England was an epiphenomenon to the process of growth. Consider that in this book there is no war, no common law, no reformation, no enclosures, no Glorious Revolution, no tariff policy, no financial system (ingeniously capital costs are derived from indexes of investment goods), and no act of union with Scotland or Ireland. In fact, there are no institutions. England is the accidental industrialist. This reader would consider the proposition that Britain is sufficiently exceptional that we may perhaps dispense with institutions or conflict after 1700. I do find it hard to understand how relative prices or technology can explain England's violently successful appropriation of a large fraction of European trade. England aside, can we write the history of Germany without the Thirty Years' War? Can we write the history of Indonesia without considering the policies of the VOC toward indigenous spice production? Can we write the history of the Americas without slavery? We cannot, and we do not because we know that everywhere except perhaps in England, economic outcomes do not simply reflect changes in relative prices but in the institutions that mediate how societies adjust to these prices. To the extent that the British Industrial Revolution was a one-off coming together of relative prices, it now belongs to the domain of the historian rather than that of the social scientist. If Allen has liberated us from the burden of the Industrial Revolution, he may have done us a great service, for we can now get on with the real problem of growth, that of the political economy of institutional change.