Manufacturing dissent

Unit­ed States man­u­fac­tur­ers, like their Aus­tralian coun­ter­parts, are indulging some hyper­bol­ic alarm about their future, but for dif­fer­ent rea­sons. U.S. eco­nom­ic growth seems too anaemic to sup­port demand in the sec­tor; Aus­trali­a’s eco­nom­ic growth seem to be bypass­ing it. Still, this self-inter­est­ed plea in the NYT from a direc­tor of GE for pub­lic sub­si­dies (“inno­va­tion sup­port”) for U.S. man­u­fac­tur­ing indus­try has a famil­iar ring. 

Rebuild­ing our man­u­fac­tur­ing capac­i­ty requires the demo­li­tion of the idea that the Unit­ed States can thrive on its ser­vice sec­tor alone. We need to cre­ate at least 20 mil­lion jobs in the next decade to off­set the effects of the reces­sion and to address our $500 bil­lion trade deficit in man­u­fac­tured goods. These prob­lems are relat­ed, giv­en that the ser­vice sec­tor accounts for only 20 per­cent of world trade.

Extract from Man­u­fac­tur­ing a Recov­ery —

Three points on which I dis­sent from this opinion. 

  1. When the U.S. econ­o­my does cre­ate net jobs, you can bet that (com­pet­i­tive) man­u­fac­tur­ing won’t be at the top of the list (or it would’t be com­pet­i­tive). It’s much more like­ly that U.S. job growth will remain dom­i­nat­ed by ser­vices industries. 
  2. The man­u­fac­tur­ing trade “deficit” is the cor­rel­a­tive of con­sumer ben­e­fit from trade; not a sign of recession. 
  3. The size of the ser­vices sec­tor in world trade is no indi­ca­tor of its con­tri­bu­tion to a man­u­fac­tur­ing econ­o­my (ser­vices trade is, in any case, huge­ly under-count­ed by cur­rent methods). 

But I agree with one of Ms Hock­field­’s sen­ti­ments; Aus­tralia, like the Unit­ed States, cer­tain­ly needs to stay in the busi­ness of real—not just invis­i­bles—out­put to sus­tain growth.

Man­u­fac­tur­ing has “shrunk” in terms of employ­ment and mea­sured-out­put (GDP) share in all OECD coun­tries over the past half-cen­tu­ry because its pro­duc­tiv­i­ty has improved, due in large mea­sure to improve­ments in pro­duc­tion and non-pro­duc­tion ser­vices inputs. But those ser­vices industries—which have absorbed labor at a rapid rate over the same half-century—may (prob­a­bly do) exhib­it low­er rates of own-pro­duc­tiv­i­ty improve­ment. The over­all out­come for the growth of the econ­o­my is the sum of the out­put growth of all indus­tries; so it’s prob­a­bly essen­tial for both (high pro­duc­tiv­i­ty) manufacturing—or farm­ing, or min­ing—and for ser­vices indus­tries that they are joint pro­duc­ers. [Geek note: I’m debas­ing a line of argu­ment from Bau­mol, to Solow to Ace­mogelu and oth­ers on the role of ser­vices in mod­ern economies. See this sur­vey paper by Joe Fran­cois and Bernard Hoeck­man or this one by Robert Inklaar et al.]

In Aus­tralia, ser­vices most valu­able con­tri­bu­tion to mea­sured-out­put may be linked to their con­tin­ued col­lab­o­ra­tion with mer­chan­dise indus­tries. We’re not like Sin­ga­pore, for exam­ple, or Bel­gium, endowed to be pre­dom­i­nant­ly a ser­vices econ­o­my based on entre­pôt activ­i­ties. The geo­graph­i­cal extent of our coun­try and it’s loca­tion means we’re endowed for oth­er kinds of pro­duc­tion where land plays a role. Cer­tain­ly, we’ve become an econ­o­my dom­i­nat­ed by ser­vices out­put, but I see zero chance that real pro­duc­tion is going away. It’s safe to pre­dict that when the price of cap­i­tal assets not spe­cialised for ser­vices pro­duc­tion (farms, mines, steel-mills) falls rel­a­tive to the price of oth­er pro­duc­tive assets there will be investors: whether from Aus­tralia or else­where is irrelevant.

Although it may, in some cir­cum­stances, make sense to sub­sidise ser­vices indus­tries (“inno­va­tion”) in the guise of sup­port­ing mer­chan­dise out­put indus­tries, what most of the evan­ge­lists for man­u­fac­tur­ing or agri­cul­ture or even resources seem to miss is that such spend­ing will very like­ly increase the size of the ser­vices input indus­try (research and devel­op­ment, logis­tics, process design, prod­uct design) and shrink the lev­el of employ­ment and pos­si­bly even the phys­i­cal scale of the out­put indus­try (as pro­duc­tiv­i­ty rises). 

Leave a Comment

Your email address will not be published. Required fields are marked *