Sunday, December 14, 2014

Culture and rationality: or why South Asian call centres can be so infuriating

I recently read Kenneth Pollack's Ph.D dissertation on The Influence of Arab culture on Arab Military Effectiveness (which later became a book). It is a very fine piece of social science which, alas, I doubt any important member of either the Obama or Bush II Administrations has read. The burden of his analysis is that Arab culture strongly militates against having effective junior offices or tactical flexibility. Arab armies can be stubborn in defence (until breakthroughs happen), and can execute meticulously planned and rehearsed offensives, showing considerable individual bravery and unit cohesion. But they are typically very bad at any sort of free-style manoeuvre, innovation or coping with the unexpected. In particular, information flows can be stunningly unreliable, as people hide failure behind false reports.

Providing the Iraqi army with lots of American equipment and training still dramatically failed in the first military test in much the way one might have expected, if anyone in either the Bush II or Obama Administrations had read Pollack's dissertation. (Particularly the discussion of how the Libyan Army was defeated by Chadian forces.)

I used to describe cultural explanations as the last refuge of the analytically bereft. I am still dubious about cultural explanations which are not properly "fleshed out" but simply "thrown at" analytical problems as sort of analytical "silly putty"--something that can be used to fit any required (analytical) hole. But, as Pollack's dissertation shows, cultural explanations--done carefully--can be enlightening.

In his conclusion, Pollack makes the following observation:
It is a peculiarly American cultural trait that we dogmatically refuse to accept the importance of culture as an influence on behavior. Only Americans could assume that all men and women are purely rational beings upon whom societal values have only minor influence. For this reason, Americans have tended to dismiss culture as a potential influence on military effectiveness. We assume that any given state will conduct its military operations in exactly the same fashion as we would because we assert that our own behavior--at least in military operations--are governed entirely purely by reason and the objective conditions of our situation, but not by cultural values. As a result, we consistently misread the capabilities and intentions of foreign powers and are baffled when they consistently conduct military operations better, worse, or just different from our own (p.764).
This failure is not so surprising. Those who migrate to the US typically do so to play the game of being American, a game which is open to anyone to play. So, Americans see folk from many different cultures (14% of American residents are foreign born) coming to be Americans and learn to discount culture; or, rather, to expect a generalised "rationality" which cannot see its own emotionality and particularity. (Contemporary scholarship on the history of emotions--Australian centre here--has been revealing how Western post-Enlightenment concepts of rationality are somewhat more emotionally based and culturally particular than is often realised: Hume's dictum that reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them has great force.) Australians are perhaps a little more inclined to see differences--if only from an awareness of not being Americans--but some of the same effects apply, as Australia is even more (proportionally, 28% of Australian residents are foreign born) full of people coming to play the game of being Australian.

Cultural exasperations
Which brings me to South Asian call centres. Companies use them because South Asian labour is much cheaper than US, Australian or other Western labour and because modern telecommunication is so cheap, that the saving in labour costs dwarfs the expense in international calls.

There are, however, some (inter-related) problems with this. One is that South Asian labour is also less productive than Western labour. Second, the folk in the call centres simply do not have the same life experiences as the Westerners they are dealing with. Third, differences in culture can lead to very unhappy customers. The notion that one just pays South Asians to do "the same job" as would Westerners falls down if, for cultural reasons, they do not see things the same way as their Western customers: i.e. they do not share a common "commercial rationality". Such as not actually seeing "the job" itself in the same way.

Indian commercial mythologist/chief belief office Devdutt Pattanaik, in the course of a TED talk on culture (transcript here), says of culture that:
Culture is a reaction to nature, and this understanding of our ancestors is transmitted generation from generation in the form of stories, symbols and rituals, which are always indifferent to rationality. And so, when you study it, you realize that different people of the world have a different understanding of the world. Different people see things differently -- different viewpoints.
He contrasts one-life cultures with many-life cultures:
Take a look. If you live only once, in one-life cultures around the world, you will see an obsession with binary logic, absolute truth, standardization, absoluteness, linear patterns in design. But if you look at cultures which have cyclical and based on infinite lives, you will see a comfort with fuzzy logic, with opinion, with contextual thinking, with everything is relative, sort of -- (Laughter) mostly. (Laughter)
Reading Kenneth Pollack's dissertation happened to coincide with both my business and a friend having very annoying experience with two Australian telco's using South Asian call centres.

Telco non-communications
Case 1:
My business is part of a complicated arrangement with Optus. Last year, we moved our office. We have found it is apparently quite difficult to tell modern corporations your business has moved. Which leads to bills not being sent to where they will be paid, so they aren't. I had got some calls from Optus, and I kept telling them I was not the person to talk to, since I did not handle those aspects of the business. One Monday, my business partner spent two hours on the phone with Optus and was assured, at the end of the conversation, everything was now fixed.

The next day (Tuesday), Optus rang me while I was driving (so I did not pick up). I found later my phone had been cut off. I rang back (on another phone, since my would not connect to Optus), getting a nice lady who could not work out what was the problem, put me on hold and said she would get someone from the Finance area to talk to me: I was on hold for a while (too long, apparently) and dropped into another call centre person, who said the account as fully paid up and re-connected my phone.

The following Monday, Optus rang me while I was driving (so I did not pick up). I found subsequently that my phone had been cut off. I was on the way to a job, so could not deal with it. I got one of my fellow presenters to text my business partner and then, at lunch time, used another presenter's phone to ring my business partner who was very aware of the problem since his wife's phone had also been cut off (being part of the same plan), this while deeply personal things were happening and friends needed to be contacted. Our office manager spent a couple of hours on the phone with Optus and was assured at the end of that that everything was fixed. She rang me and told me that if I powered off my phone, it should be fine when I powered it back up. Verily, this was so.

The next day (Tuesday) Optus rang me. For once, I was not driving. We had a pointed discussion about telling them not to ring me and that we had been assured everything was fixed. I gave him my business partner's phone number, who later texted me back that he had a "pointed discussion" where he was assured everything was fixed (by my count the fourth such assurance in a little over a week) and he assured them that, if there was any recurrence, we were going to the Ombudsman.

So, our experience with a telecommunications company that apparently cannot communicate or handle communication. How much cultural issues played in the recurring screw-up, hard to say.

Case 2
Meanwhile, a friend of mine was moving between states. She rang her telecommunications company, Telstra, to tell them she was moving (and gave them the dates) and enquire about getting internet at her new (non-metropolitan) address. She later found her internet had been cut off (extremely inconvenient given how much information searching, ticket arranging, etc one nowadays does over the net). She rang Telstra and was told that it had been cut off because she was moving. She informed them that yes, but she had given them the dates and still needed internet access. After bouncing around South Asians who were not helpful, she finally got an Australian lady who reconnected her.

She subsequently found that, once again, her internet had been cut off. After more bouncing around South Asians at a call centre, where she had to explain the problem, over and over again, she finally got the same Australian lady who expressed puzzlement about what folk (in Telstra) thought they were doing, and reconnected her.  So, in the middle of packing and arranging an inter-state move, she spent hours on the phone simply because her telco couldn't apparently cope with the idea of moving at a date they had been informed of. Another telecommunications company that apparently cannot handle communication.

But, of course, communication is more difficult when people do not have the same life experiences; not merely individually, but also collectively. (I am guessing that Indians moving probably do not do very much arranging matters over the internet.) What my friend found added an extra level of annoyance was that the call centre folk she talked to did not use the language of responsibility; she got no sense at all that they thought of themselves as representatives or agents of Telstra. She found talking to the Australian lady much more satisfactory, as she used the language of being responsible representative and then acted on the same.

Culture matters
But that is also a matter of life experience and culture; the things that affect the way you see the world and think about it, other people and yourself in relation to same. Devdutt Pattanaik again:
Indian music, for example, does not have the concept of harmony. There is no orchestra conductor. There is one performer standing there, and everybody follows. And you can never replicate that performance twice. It is not about documentation and contract. It's about conversation and faith. It's not about compliance. It's about setting, getting the job done, by bending or breaking the rules -- just look at your Indian people around here, you'll see them smile; they know what it is. (Laughter) And then look at people who have done business in India, you'll see the exasperation on their faces. (Laughter) (Applause)
Remembering that even the concept of "the job" can be culturally specific. Kenneth Pollack defines culture as:
as the set of learned, shared values, patterns of behavior, and cognitive processes, developed by a community over the course of its history. ... it is acquired behavior, learned by members of the community over the course of their lives (p.38).
Culture matters for individual behaviour because:
Culture influences an individual's preferences and priorities. By defining what the individual is likely to consider important, culture shapes an individual's preferred outcome in a given situation. ... Similarly, culture will shape the courses of action and methods an individual is predisposed to employ to secure a goal. Culture has a tendency to suggest that certain ways of doing things are better than others, thus culture shapes both ends and means. Finally, culture may actually shape the way in which an individual thinks and how he or she approaches different situations.
In addition to its impact on the individual, culture also influences the behavior of groups by shaping interpersonal behavior. It teaches members of a society how to treat other people and how the individual should behave when part of a group. It establishes what is permissible and what is desirable behavior in public or within smaller groups (Pp38-9).
So culture establishes tendencies in behaviour rather than rigidly determining individual actions; tendencies that, as Pollack points out, will be clearer the larger the groups of people, and the longer the time frame, being examined.

Devdutt Pattanaik talks of one-life versus many-life cultures. But one can also talk of limited versus generalised morality, nicely defined by economists Avner Greif and Guido Tabellini as clan versus city (pdf): 
In a clan, moral obligations are stronger but are limited in scope, as they apply only toward kin. In a city, moral obligations are generalized towards all citizens irrespective of lineage, but they are weaker, as identification is more difficult in a larger and more heterogeneous group.
Christianity and Buddhism encourage generalised morality, Confucianism and Hinduism encourage limited morality. Islam is a limited morality with universalised ambit. What folk in the West think of as "common sense" rests to a significant degree on generalised morality and putting yourself in the other person's situation. Limited morality cultures tend to not have the same "common sense". 

A way to think about culture is a set of economising heuristics where the alignment of one's perspectives, expectations and preferences with those of others decreases cognitive effort and reduces social friction (i.e. transaction costs). The more said heuristics align with social success (marriage, income, reputation; i.e. have strategic complementarity [pdf]) the more they will be reinforced, and so persevere. Conversely, persistent and socially significant shifts in pay-offs will lead cultures to change (as they do). 

So, you hire folk who continue to live in their culture to deal with practical and personal aspects of people's lives who live in a quite different culture with quite different collective life experiences. How is that going to work out, in the customer satisfaction stakes?

Helping to create telecommunication companies that handle communication really quite badly, apparently. 


[Cross-posted at Skepticlawyer.]

Thursday, December 11, 2014

Domain states, tax states and fiscal states

Have been reading a Ph.D dissertation by Wenkai He on the consolidation of a modern fiscal state in C18th England and late C19th Japan and the failure to do so in late Qing China. (The dissertation has since become a book.) The dissertation references history of public finance literature I was previously largely unaware of. The literature covers domain states and fiscal states (what Joseph Schumpeter apparently called tax states). Wenkai He further divides the fiscal state into the traditional and the modern fiscal state. I would prefer to talk of domain states, tax states and fiscal states, defined as follows:
The domain state: draws income primarily from state-owned property or enterprises such as estates, forests, mines, railways, monopolies, factories, etc. 
The tax state: draws income primarily from taxing economic activity, using its tax revenue mainly as income to meet its current spending, with any borrowing limited to short-term occasional expedients. 
The fiscal state: draws income primarily from taxing economic activity, using its tax revenue as a means to leverage more financial resources, either through issue of paper notes or through long-term borrowings or both.
In other words, fiscal states both rely primarily on taxing economic activity and can deficit finance in a significant and recurring way. The first fiscal states were Song China, the first state to issue paper notes, and the Serene Republic of Venice, which issued the first public bond, the prestiti.

There are no longer any tax states. It was an historical form which, with the universal adoption of paper/plastic notes and development of global capital markets, is now extinct. The Roman Empire, the late Sassanian Empire (at least after the reforms of Khosrau I), the Abbasid Caliphate, the Ottoman Empire, were all tax states. As were Ming and Qing China, as the large scale use of paper notes lapsed after the experience of hyperinflation at the end of the Yuan dynasty and of inflation in paper notes at the beginning of the Ming Dynasty.

Ruling domains
The domain state was the original form of the state. Pharaonic Egypt, Mesopotamian palace-and-temple complex states, the Khmer Empire, were all domain states. But so was the Spanish Empire after the discovery of the Potosi "silver mountain" and other silver mines. So was the Soviet Union. So are Saudi Arabia, Iran and Venezuela. So are Cuba and North Korea. The domain state is far from extinct.

The odd thing is, I had come to various conclusions about domain states without being aware of the term, though having the term is usefully clarifying.

Domain states come in two basic types; those reliant on a single resource (silver, oil) and those reliant on the extraction of labour surplus (agrarian and industrial domain states).

Both forms are [likely to be]* highly ideological and are the more ideological the more the state dominates economic activity. That is, both have a set of ideas that express the status and dominance of the ruling authority, with public adherence to said ideas signalling loyalty to the authority that is economically dominant (and thus has very extensive power to reward or punish by giving access to, or cutting off access to, the revenues it controls). While cults of personality provide a way to signal loyalty when said loyalty is effectively compulsory. Failure to signal loyalty in the required way is likely to make one a target; a useful and public target for signalling the error of not publicly adhering to the status and dominance game of the ruling authority. Said status and dominance being all the stronger as the ruling authority--due to the scale and independence of its revenue--typically has little need to conciliate or bargain with other sectors in the society.

So, if the domain state is religious, it is likely to be very religious--hence the persistence of the Spanish Inquisition and the particularly virulent religious police of Saudi Arabia and Iran. If secular, it is likely to be intensely ideological--as in the Chavismo of Venezuela, the Castroism of Cuba, the Juche of North Korea. Or the Leninism and Stalinism of the Soviet Union, the Maoism of the People's Republic.

Extracting surplus
Domain states reliant on the extraction of labour surplus do so by paying labour well below its marginal product. Pre-industrial (i.e. agrarian) domain states typically did this by tying peasants to the land, although extraction through labour service or corvĂ©e was also often used. Hence agrarian domain states are typically addicted to monumentalism: labour service cannot be "stored" from one year to the next, thus it has a "use it or lose it" quality. So a continual series of autocrat-controlled labour service projects signals, manifests and preserves the authority of the autocrat; especially as the monuments themselves typically represent in stone the status and dominance of the autocrat, either directly or by expressing the status-and-dominance symbology.
Jayavarman VII likely used his face on his monumental temple.

Industrial domain states extract labour surplus by controlling the borders and blocking exit. Since the state owns all the enterprises, it can ensure that workers are paid below their marginal product. Which means they can clearly do better elsewhere, hence they have to be blocked from leaving so that they can be paid well below their marginal product. It is exactly the same principle as serfs being bound to the land (whether Spartan helots, late Roman coloni or medieval serfs), except on a much vaster scale.

Which is why, when former domain states transition to fiscal states, they open their borders, even if the same regime remains in power (as in China and Vietnam). No longer reliant on paying people below their marginal product, they are no longer anywhere near as motivated to block exit but very keen to encourage transactions that can be taxed--including cross-border transactions. So, Leninist regimes running fiscal states open their borders, while Leninist regimes running (industrial) domain states close them.

States making societies
The state is not an epiphenomenon of society, but a fundamental moulder of it. States make societies, and societies make states. And the more the former is true, the less the latter is.

Stateless societies either pre-date the state or are created in refuge from it. An option which is no longer open. (Sufficiently cheap space travel--if that is possible--may recreate it; but not if the experience of European colonialism is any guide.)

But the tax, and later the fiscal state, has an interest in encouraging transactions to be taxed. Which (along with various factors in organising military force) eventually led, in Europe, to the creation of the bargaining state. Which was so successful, the idea developed that states were an epiphenomenon of society. But that was a false inference from profoundly anomalous experiences in profoundly unusual (in terms of human history) societies.

It is one of the great historical ironies, that the philosophy (Marxism) which most expressed this notion that the state was an epiphenomenon of society--that it reflected and served the interests of the "ruling class"--in its Leninist variant expressed most completely how false this notion was, as Leninist states created the societies they wanted. Not wanted in the sense of some perfect ideological vision, but wanted to serve the interests of the ruling authority of the state. It was control of the state which made them the ruling authority, they did not control the state because they ruled. Their state made their society. Up to the limits of rival states.

The emergence of the state was the emergence of surplus-extracting hierarchy able to mould society into a form that could support it. States originally made societies rather more than societies made states. Since then, the degree to which states made societies, or societies made states, has been highly varied; but states have never been a epiphenomena of society.


* [Small city-states not so much, since they have a de facto cosmopolitanism more or less imposed on them.]
[Cross-posted at Skepticlawyer.]