I was a bit slow to respond to Kieran’s post on the World City System, but let me say that my views on this system are pretty much a cross between Wired and William Cobbett. In a world where nearly all legitimate work of high-pay and status can be performed electronically and remotely, the most plausible explanation of ‘global cities’ is that they facilitate cronyism and corruption.
Updated with a little more evidence 25/2
On this point, Virginia Postrel has an interesting piece (abstract free, full piece payment required) on the work of Chicago economists Rajan and Zingales saving capitalism from the capitalists. Essentially, their claim is that where that where finance is allocated on the basis of personal relationships, it becomes a tool for creating and protecting monopoly. This is what they call “relationship capitalism”. Others have used the more pejorative phrase “crony capitalism”.
Postrel uses these ideas to attack the idealised, and largely mythical, small-town bankers of the past in favor of today’s more impersonal system. It’s certainly true for retail borrowers that relationships with bankers are no longer important. But she misses the irony that while distancing themselves from most of their customers, members of the financial sector have gathered together ever more closely in centres like New York and London.
Similarly, this Buttonwood column from the Economist deplores the fact that house prices in London are being bid up by City types who, he suggests, have enriched themselves at the expense of their customers (he’s referring to the mutual funds scandals, but these are just the latest of many). He doesn’t, however, ask the obvious question: Why do these City types crowd together in London (and New York). After all, the same City types are busy telling us about a globalised world, linked instantaneously by the Internet. And, as Warren Buffett has shown, they are right. You can get all the information you need to formulate market-beating investment strategies while sitting in Omaha, Nebraska.
The two halves of Buttonwood’s observation are linked by the much older observation of Adam Smith (quoting from memory here)
Men of the same trade seldom gather together, even for innocent merriment, but the meeting ends in some conspiracy against the public.
The work that financial institutions are supposed to perform, trading assets and allocating risk in transparent markets, can be done anywhere on the planet. It’s the stuff they want to do without any inconvenient records, and with the kind of trust that’s needed for conspiracy that requires clustering in a central location where social bonds can be cemented by eating, drinking and sleeping together.
I should add (and at this point, readers might want to note that I am located in Brisbane, Australia, which may engender a somewhat jaundiced viewpoint) that most of what I’ve said above applies to academia. On balance, the clustering of high-status academics in places like Oxford, Harvard and Chicago has consequences that are more negative than positive. The merits of intensive collaboration and casual hallway discussions of academic topics are more than offset by the clubbishness and mutual backscratching produced by these concentrations.
Update 25 /2Another piece of evidence on this
Gains from Corporate Headquarters Relocations: Evidence from the Stock Market,” Journal of Urban Economics, Vol. 38(3), November 1995, 291-311, Chinmoy Ghosh, Mauricio Rodriguez, and C. F. Sirmans. This paper provides empirical evidence on investors’ perceptions of the relative advantages and costs of spatial agglomeration. Specifically, we examine the stock price effects of headquarters relocations. The stock market reaction is significantly positive when relocation decisions are attributed to cost savings, indicating that cost savings available at less centralized locations outweigh any loss of enhancements associated with spatial clustering at urban centers. In contrast, decisions prompted by managerial self-interest and desire for luxurious offices elicit an adverse reaction from investors.(emphasis added)
You can download the paper here (2.8 MB Word doc).
{ 18 comments }
Conspirator 02.24.04 at 10:40 am
The work that financial institutions are supposed to perform, trading assets and allocating risk in transparent markets, can be done anywhere on the planet. It’s the stuff they want to do without any inconvenient records, and with the kind of trust that’s needed for conspiracy that requires clustering in a central location where social bonds can be cemented by eating, drinking and sleeping together.
It is when you skip from theory to things like conspiracy that I’m starting to have doubts about it.
For example, a dynamic labour market is a benefit both to the employer and the employee. And the concentration of these companies in small geographical areas creates such a market. And indeed the this market is oiled by ‘eating, drinking and sleeping together’. But this market is no conspiracy.
Chris Bertram 02.24.04 at 11:40 am
_In a world where nearly all legitimate work of high-pay and status can be performed electronically and remotely… _
But this just isn’t true. See ch. 3 of Brown and Duguid’s _The Social Life of Information_ for some discussion of the downsides of people doing their work isolated from one another at the end of an telephone or computer terminal.
Management of people, at least good management of people, requires face to face contact (at least sometimes). And those aspects of the day-to-day running of, say, an academic department, that get achieved over lunch or in the corridor aren’t best represented as “conspiracy” (!).
Smith’s claim is that social intercourse facilitates and encourages conspiracy. Your claim is that the only or chief explanation of social intercourse among those of a similar trade is conspiracy. Those aren’t equivalent, and there are plenty of good non-conspiratorial reasons for clustering together.
Greg Hunter 02.24.04 at 12:45 pm
There is no doubt that wealth concentration appears to be occurring primarily on the east/west coast in the USA. Is that a result of conspiracy or unintended consequences? I suspect a little of each.
I look at it from a Midwest perspective and over the last 50 years the companies that were founded in this area and held privately were ultimately sold to investors through the vaunted stock exchange (NY). The offspring of these rich individuals have a tendency to flee the area that helped make them and take their capital in tow. Because there were so few, relatively, wealthy people in the Midwest, the people of wealth seem to concentrate together, because they have similar experiences/beliefs. This concentration tends happen in NY, FL and MA for Dayton, Ohio. If you happen to make your cash in the Entertainment and to some degree the computer field, your capital, and your ass, would head to Cali. This cycle will continue as the process that gave you capital can be invested so that you never have to return to your former address. Your kids will go to private schools and mix with the genetic offspring of people with similar heritage. The relationships form in the East will nurture your career all of your life as your friends will be come stockbrokers, CEO’s, CFO’s and these casual relationships in the seats of power, will increase wealth in those seats of power. The view of the east coaster tends to believe that if you do not come to the east and become successful, you really are not playing against any competition, therefore you are not worthy of consideration.
People from the East tend to think of the Mid West as a third world country. Is this a result of our best and brightest leaving the area, along with their capital, to consolidate their wealth with NY? Hey, but I can look on the bright side, while people of culture get new museums, plays, art and a place on the world stage, the Midwest gets a new nuclear processing plant and a prison.
Thanks for giving us, really, all we can handle.
Chirag Kasbekar 02.24.04 at 2:53 pm
I also suspect there is some path dependence at work here. After all, these clusters were usually formed — and the financial firms were located in them — much before the internet came along, right?
paul 02.24.04 at 3:00 pm
<quote>Management of people, at least good management of people, requires face to face contact (at least sometimes). </quote>
You manage things, but you lead people.
<quote>And those aspects of the day-to-day running of, say, an academic department, that get achieved over lunch or in the corridor aren’t best represented as “conspiracy†(!).</quote>
The relationships of the people involved is key here: an associate professor and the head of his department taking lunch is far different from the deans or two or more schools meeting. I think the quote from Smith was pointing out the risk of collusion when enough of the dominant players in a market get together.
The whole notion of “location, location, location” and the death of distance is something I have been taking note of: living in a region defined by pronounced boom and bust cycles, I take a personal interest in it.
Chirag Kasbekar 02.24.04 at 3:06 pm
I agree with Chris that the potential for conyism is probably just one of the reasons.
I found this on London:
http://www.cityoflondon.gov.uk/business_city/research_statistics/pdf/clustering_exec_sum.pdf
pw 02.24.04 at 5:30 pm
Chris Bertram doesn’t go nearly far enough. It’s not just that many kinds of work can’t be done effectively at a distance, but also that when you restrict the universe of “interesting” work to those things that can apparently be done effectively at a distance, you’ve given the organization a case of ALS.
Meanwhile, it’s quite possible that a mixed economy of face-to-face and low-latency remote interactons really isn’t viable in the long term, any more than a mixed economy of cash crops and food farming.
John Quiggin 02.24.04 at 7:14 pm
I’ve updated the post with the following piece of evidence, which I should have chased down before posting rather than after.
Gains from Corporate Headquarters Relocations: Evidence from the Stock Market,†Journal of Urban Economics, Vol. 38(3), November 1995, 291-311, Chinmoy Ghosh, Mauricio Rodriguez, and C. F. Sirmans. This paper provides empirical evidence on investors’ perceptions of the relative advantages and costs of spatial agglomeration. Specifically, we examine the stock price effects of headquarters relocations. The stock market reaction is significantly positive when relocation decisions are attributed to cost savings, indicating that cost savings available at less centralized locations outweigh any loss of enhancements associated with spatial clustering at urban centers. In contrast, decisions prompted by managerial self-interest and desire for luxurious offices elicit an adverse reaction from investors.(emphasis added)
Chris Bertram 02.24.04 at 7:31 pm
I’m struggling a little to see the relevance of your last comment to the broader question. The fact that the stock market reacts positively to a cost-saving relocation decision is unsurprising, especially given short time-horizons.
John Quiggin 02.24.04 at 7:49 pm
In general, stock markets react favorably to investment projects if they expect positive returns (in NPV terms) from those investments. Even if justified in terms of cost-cutting, a reduction in, say, R&D expenditure won’t in general, produce a positive stock market reaction.
Considered as an investment, a relocation to say New York produces a stream of [putative] benefits (those of proximity and spatial agglomeration) and costs (higher rents which raise costs both directly and through the need to pay higher salaries). The stock market reaction says that, in the view of investors, the costs outweigh the benefits.
More precisely, the adverse stock market reaction to moves to global cities suggests that any net benefit is less than the initial cost of relocation. But the stronger result is implied by the positive reaction to relocations to cheaper locations.
From what I can see from the paper, previous studies have found much the same thing.
Chirag Kasbekar 02.24.04 at 8:53 pm
For some reason, I thought the post was by Henry!
But John, I’m probably missing something (it’s late in the night here — early in the morning, actually), but does that paper say anything about how benefits from clustering are got?
(Though the ‘clubbiness’ that the Corp of London report actually touts is probably a pretty significant reason.)
From the exec summ of the Corporation of London report:
“In the sphere of financial services, the research informs us that large, medium and small-sized financial service firms have a tendency to cluster in metropolitan areas because of the need
to: access large pools of specialist labour and support services (e.g. accounting, actuarial,
legal etc.); be in close proximity to the markets; benefit from agglomeration economies,
which reduce transactions costs; develop and innovate intrinsic skills through the sharing of
knowledge and practice (Davies, 1990; Roberts et al., 2000). Financial service firms that locate in strong clusters grow faster than average and strong financial clusters attract a disproportionate volume of new firm entry (Pandit et. al., 2001).”
Another major reason the report cites is branding.
BTW, the whole report can be got from here:
http://www.cityoflondon.gov.uk/business_city/research_statistics/research_publications.htm
John Quiggin 02.24.04 at 9:38 pm
You can download the paper I referred to
here (2.8 MB Word doc).
Chris Bertram 02.24.04 at 10:13 pm
John,
I think your Warren Buffet point was a bit misleading since corporate relocation raises different issues from, say, homeworking.
But even if investors would like Acme Bank Inc to relocate to Omaha, Nebraska on cost grounds, where’s the evidence that the people Acme Bank needs to work for them and to make those decisions are happy living there? After all, cities do a lot more for people than provide locations for enterprises. I doubt that Omaha, Nebraska has many decent restaurants and I’ve never heard of their opera house … Presumably if Acme Bank moves to Nebraska and Acmecompetitor Bank stays in New York and takes all their best people, Acme Bank is screwed (or am I missing something).
John Quiggin 02.24.04 at 10:31 pm
If investors are rational, they will take these effects into account when the mark a company up or down for relocation decisions. After all, the cost savings would be illusory if the company could not hire good workers for the same wages in the new location.
The test question here is whether wages for workers in a given occupation and with comparable skills are higher in NY or in competing non-global-city locations. If workers value the opera more than they dislike high rents, wages in NY should be lower. My perception is that the opposite is true, but the proviso about comparable skills is important and hard to check.
If investors aren’t rational, are characterized by short-term bias etc, then there’s no general reason to think that location decisions will be optimal. In particular, if investors aren’t fully rational, the constraints on managers pursuit of their own self-interest will be weaker. I think, for the reasons you mention, it’s clearly in the self-interest of managers to locate in NY and other global cities.
Note that high wages for skilled staff benefit managers, in view of the general corporate rule that the supervisor must be paid more than the supervised.
dsquared 02.25.04 at 1:24 pm
I have a number of views on this, but since I’m a financial professional working in a major cluster, I suppose I’d better be careful what I say!
Chris Bertram 02.25.04 at 2:47 pm
Unless you want to end up living in Nebraska …
David Foster 02.25.04 at 3:52 pm
Geographical clustering happens in all sorts of industries, not just finance. People in firm “A” decide to go off on their own and start firm “B”; they don’t want to move (and want to be able to conveniently recruit former associates), so they put it in the same city. The same happens when people from “B” go off and start “C”. In some industries, the process is further encouraged by the presence of local suppliers and support companies.
Ian Whitchurch 02.27.04 at 11:15 pm
This is interesting for me, because I am at the pointed end of all this.
Despite my .uk address, I live in Canberra, and I am working on taking a Perth-based oil company public on the Sydney-based Australian Stock Exchange.
Our CEO is in Perth, our board comes from Perth, Sydney and Melbourne, and our assets are in South Australia and the Northern Territory.
Needless to say, I’ve been doing a lot of travelling lately.
An example of the issues involved with distance is our dealings with a Swiss-based brokerage house with offices in Sydney and Melbourne.
In Australia, they are mostly based out of Melbourne, but their oil and gas guy is in their subsiduary Sydney office.
Thus, we first went to Sydney to sell their oil and gas guy on our company, and he likes us.
But as he was based in Sydney, he never got to talk to Melbourne face-to-face, and that meant that the Melbourne people were less enthusiastic about the company than they could have been.
Would this have been different if they had all been working out of the one building, or at least the one city ?
My guess is “Probably”.
My reply to Adam Smith’s comment is that trade is about relationships, and that meeting people face-to-face and socially helps build trust relationships – there is stuff you just cant do over the phone or via the net.
Ian Whitchurch
PS D-squared, or anyone else involved in the finance scheme – can I talk to you about the AIM, and it’s possible role with a Brisbane-based oiler that controls (among other assets) the oil lease for the Falkland Islands ?
Comments on this entry are closed.