If I may be so bold as to intervene in this discussion -- 1. To demonstrate a significant shift it should not be necessary to prove that virtually all, or nearly all work in a category has been "globalized." 2. It might be helpful to distinguish between personal and businesses services. 3. The issue of markets is important, but let's not confuse market focus with how or where a product or service is produced. 4. Regarding language, I can't recall the source, but somewhere in the last couple of years I seem to recall that publishing companies in the U.S. have taken to shipping their manuscripts to Asia for input by workers who cannot read a word of English. For some time American Airlines (and others) have shipped their ticket data input to Barbados. 5. With respect to semiconductor mfg., leading edge companies are setting up parallel computer controlled mfg. facilities around the U.S. and in Europe, the Paicifc Islands, and Asia. Chips are designed and initial prototype mfg. runs are done in the Silicon Valley. Once reliable yields are achieved, mfg. is transferred to satellite plants where identical production lines are programmed to mfg. in quantity. All that need be transferred is the computer instruction and a few engineers to monitor output. This was not possible even ten years ago. 6. It would be helpful to take a more differentiated view. Not all industries are equal with respect to their vulnerability to globalization. Tortilla mfg. in LA can't be relocated because the market requires and demands daily delivery of fresh product. Pool cleaners can be recruited anywhere but they must be physically located where the pools are. Generally products or services that lend themselves to computerization or involve information management are more vulnerable. Travel agencies can now subcontract their phone sales and bookings to prison labor across the nation. Globalization is an uneven process affecting different sectors and different segments of the labor market in different ways. 7. Globalization is not caused by but is facilitated by increasing computer power and the declining power/price ratios made possible by each new generation of microprocessor. Digital technology makes possible reengineering (re-Taylorization) of complex tasks and skilled work that previously could be performed only by highly trained workers. The extent and penetration of this process will be accelerated as elements of the work process can be subjected to electronic execution or control. This is likely to take a quantum leap once Artificial Intelligence capabilities are refined and integrated into production processes, allowing computer programs to appropriate knowledge and discretion now held by skilled workers. 8. Never underestimate the vulnerability of a complex system to meltdown ("normal accident"). No computer program written has ever outperformed the capacity for flexible judgement of the trained and experienced human brain. The most powerful use of digital technologies is to augment, not replace, the worker (an argument made persuasively some years ago by David Noble and generally ignored by profit-driven short horizon management). Well, that's about two cents worth of opinion. So I'll stop there. Michael. At 11:40 PM 5/7/97 -0700, Tavis Barr wrote: > >[I wear the title proudly. :) ] > >On Wed, 7 May 1997, D Shniad wrote: > >> Tavis: >> >> My contention is that service markets aren't as globalizable as >> manufacturing markets. >> >> Sid: >> >> I don't think this is anything more than a contention. > >Allright, I'll use some data. It comes from a data set I'm working on >that studies employers in four different areas (New York, Kalamazoo, >Philadelphia, Pittsburgh). Admittedly it might be a non-random sample >geographically, but it's what I have handy. The question is "What is >the main market for your firm's goods or services?" and the answers >(weighted by firm size and to correct for industry oversampling) are: > >The Neighborhood 40.2% >Metropolitan Area 42.1 >National 10.1 >International 7.6 >[N=316] > >For those in the service sector (SIC1=9), the answers are: > >The Neighborhood 34.9% >Metropolitan Area 54.7 >National 8.5 >International 1.9 >[N=106] > >I don't have a big enough sample in manufacturing and telecommunications >to get consistent results. :( There must be some big national marketing >survey somewhere that has a similar question, though.... > > >> Tavis: >> >> The same types of jobs you describe in telecommunications (operator >> services) have their analogues in many other sectors: Claims processing >> in health care, credit card and loan processing in banking, catalog sales >> in retail trade. They can all be moved anywhere around the country >> (though I suspect that language difficulties at least would make it hard >> to move them across borders). >> >> Sid: >> >> The work doesn't necessarily involve speaking. If it does, English is spoken >> by folks across Canada, the U.S. and Britain, as well as by a great many >> educated folk around the world; Spanish is spoken in Europe, Mexico and >> South America, as well as across the U.S., etc. > >C'mon, that's stretching it. Would you trust non-Latino American clerical >workers to process forms in Spanish? > >> But the key thing is that a great of this work is computer-based and doesn't >> involve speaking at all. The maquila-based postal sorting (by Spanish- >> speaking workers sorting mail that's located in Chicago) is prototypical of >> what I'm describing. By the same token, remote trouble analysis of the >> phone system can be done in the same locale as where the trouble is or >> across the world. >> >> It's not just low end jobs that we're talking about, either. The Indian >> software industry is state-of-the-art in sophistication, but it pays wages that >> are a fraction of those paid software writers in Europe and North America. > >This just sounds like a column out of _Wired_ magazine. I could continue >it and write up all these wonderful examples of the neat brain jobs that >are being created by computerization and all of the menial jobs that >they're replacing and how production is obsolete and the new economy is >information based. But gee whiz stories are just that. > > >> This is what makes Robert Reich's promotion of education, training and >> skills such a meaningless response to capital's restructuring of work; capital >> doesn't have to choose between high skills and high wages on the one hand >> and low skills and low wages on the other. It can have *both*. > >Yes, yes, yes, but what does that have to do with globalization? The >same is true of the City University of New York replacing all its faculty >with low-paid gradutate students, most of whom are from the City. > >> Tavis: >> >> However most jobs in health care (nursing and assistance, hospital clerical >> and mainaintance work), banking (teller and sales work), and retail trade >> (person-to-person sales, service) are much harder to move very far. >> >> Sid: >> >> You're over generalizing here, Tavis. Hospital clerical work can be handled >> in exactly the same manner as what I described in the phone industry. It >> can easily be provided at the end of a phone line or by forcing people >> signing in or out of hospital to provide their personal information via a >> computer screen. Banking tellers are going the way of the dodo bird. At >> least in Canada, their numbers are dropping precipitately (I doubt this is >> unique to this country) as their ranks are replaced by computerized tellers >> and ATMs. > >You're over exaggerating, Sid. When was the last time you checked into a >hospital? (Me, last October when I made a boboo with a utility knife and >nearly sliced my thumb off). I've temped at many clerical jobs in >hospitals. They consist of doing lots of bullshit for doctors (taking >dictations, making tennis dates), xeroxing things to put in files, >keeping track of patient rosters; I also had a job for a few months as the >data entry clerk in inventory for a big hospital in Boston (Beth Israel). >I'm sure many of the tasks (requisition forms, patient logs) could be >slightly automated, with due thanks from a lot of clerical workers. But >most of the work is sort of informational clean-up, and depends on >interaction with doctors, nurses and patients to figure out what the >hell is going on. Remember also that medical files are generally kept >off-line and decentralized for privacy reasons. > >As to the Teller, the numbers are certainly falling, though they haven't >disappeared. I extracted this from the CPS -- the latest year available >is 1995. It's the percent who self-identified as tellers, estimator SE in >parenthesis (the SE's are similar because it's a binomial distribution): > >83 .00444 (.00023) >84 .00420 (.00023) >85 .00445 (.00023) >86 .00397 (.00022) >87 .00040 (.00022) >88 .00432 (.00023) >89 .00419 (.00023) >90 .00410 (.00022) >91 .00428 (.00023) >92 .00454 (.00024) >93 .00349 (.00021) >94 .00352 (.00021) >95 .00368 (.00022) > >I mean ya they're down a bit but at this rate it'll take several decades >for them to disappear. On the other hand, there are more than enough >nurses, nurses aides and orderlies to make up for them (the latter, in >NYC, icreasingly being workfare workers): > >83 .03326 (.00063) >84 .03263 (.00063) >85 .03317 (.00062) >86 .03338 (.00063) >87 .03388 (.00064) >88 .03379 (.00063) >89 .03401 (.00066) >90 .03295 (.00062) >91 .03313 (.00062) >92 .03526 (.00065) >93 .03756 (.00067) >94 .03904 (.00070) >95 .03901 (.00070) > > >> You keep pushing examples of face-to-face contact. But the point is that >> this is exactly what is being targeted and reduced, replaced by the use of >> computerized services. > >They're certainly trying. But then they've been trying for over a >century to get rid of workers entirely and have factories that are just >run by machines. Maybe someday they will (oh, did I tell you the one >about my old roommate's friend who runs a middle eastern bread bakery in >Boston? His bread is all made by computer and his only employees are >delivery drivers. How about the kibbutz that makes irrigation equipment >and all of the employees are computer technicians overseeing control >specs on the machine tools?). But a few gee-whiz anecdotes don't mean >we're there yet. > > >Cheers, >Tavis > >