Wednesday, September 23, 2009

Singular Priorities

In a procession, in the line of succession to the throne, in the sequence of steps that form the critical path to complete efficiently and rapidly a course of action, a person/pretender/step necessarily takes precedence over all others in a pre-assigned ranking. In this one and only sense one can discuss the set of their priorities, for there is no trade-off between the inclusion of one person/pretender/step and the inclusion of others. Fine. And you can say, with the Mahatma Ghandi, that “Action expresses priorities”, because the ranking does not precede but follows the actual decision and there is no longer a trade-off. Or, with Steven R. Covey, "Priority is a function of context": this is precisely the point.

The trouble is that people in general, and in particular politicians and bureaucrats – and especially New Labour politicians and Brussels eurocrats – talk incessantly and over-abundantly about “priorities” in the sense of ranked policy-objectives, in situations in which there is a trade-off between them, both in the terms in which their relative achievement is feasible, and in the terms in which they may be regarded as substitutes by the decision-makers. This habit is sloppy, meaningless and misleading; it should be forbidden, penalized and eradicated.

Yet examples abound. A Google search gives about 34,100,000 entries for priorities, 10,100,000 for top priority, 21,100,000 for Obama’s top priorities for 2009. “Brown promises 'new Government, new priorities'” (27 June 2007). Tony Blair: “It is not an arrogant government that chooses priorities, it's an irresponsible government that fails to choose." Peter Mandelson: “Europe has to address people's needs directly and reflect their priorities, not our own preoccupations.” “This page [on President Barroso’s website] gives direct access to all information regarding the Commission's current priorities”. Angela Merkel: “You should know what our priorities are.” And again Merkel: “I think it is a right idea to stage a special summit, which would deal with the question of priorities of European politics;” and again: “It must be clear what the priorities on the agenda are.” Kofi Annan: “From this vision of the role of the United Nations in the next century flow three key priorities for the future: eradicating poverty, preventing conflict and promoting democracy.” John Berger: “It is not, as poverty was before, the result of natural scarcity, but of a set of priorities imposed upon the rest of the world by the rich”. US ex-Vice President Dan Quayle: "For NASA, space is still a high priority". Peter MacKay: "What Canada has to do is to have a government connected to the priorities of the people of which it is elected to serve. Those priorities include ensuring medicare is sustainable, support for the military, and tax and justice systems that work". Bill Clinton: “It's the American people who sent us here; it's our obligation to meet their priorities. So let's roll up our sleeves, get back to work and finish the work we were sent here to do".

“Di mamma ce n’√® una sola” is an old Italian saying: you only have one mother. In the Roman Law sense of “mater semper certa” this has been falsified by progress with artificial insemination and wombs for rental. But strictly speaking, and in the sense of a unique emotional link with one special person, the saying remains true. My old friend Gemma, fed up with her little daughter’s apparent lack of respect, used to remind her adding: “Tutte le altre sono matrigne” (All the others are stepmothers). This is why I teach my students that priority is like your mamma: you can only have one. All the others are objectives, or targets - either independent of each other and therefore simultaneously pursuable without their having to be ranked, or impossible to rank without reference to 1) their actual trade-off with one another, in comparison with 2) the trade-off preferred by the decision maker, individual or government. Thus by definition you cannot have priorities, let alone top priorities; as you can have only one, that one can only be top, so the top is redundant.

A philosopher friend tells me that I am too literally minded: “Priorities” - she says - simply mean a set of objectives that are singled out as generally desirable, as opposed to all the other possible objectives which are regarded as less desirable, or indifferent, or undesirable outright. But I am adamant and unrepentant. If we have a list of undesirable things, their reduction or cessation or elimination must be regarded as desirable. Therefore ultimately something is either indifferent, or desirable in some form (including its opposite when it comes to undesirables). And the desirability of desirables cannot be stated in abstract, but is exclusively subject to its rate of substitution with other desirables, in the decision-maker’s system of preferences, being greater than the rate at which the desirable in question can be obtained at the cost of another desirable. Cost can also be measured in terms of a common desirable such as money, or the time and/or effort it takes to explore the desirability and substitutability of a desirable objective for another. There is no point in stating that something is desirable, without specifying up to what point it is desirable, either in terms of its direst cost in terms of money, time and effort, or in terms of the lower degree of achievement of other desirable objectives.

Anyone listing as alleged priorities a number of conflicting objectives, whether ranked or unranked, has said absolutely nothing about their relative desirability and the extent to which each one can/should be sacrified to the higher realization of another. This is what most of microeconomics is about: the comparison of the rate of transformation of one good into another in production and in the system of preference of a decision-maker, ideally both equalized to relative prices in the market. In the most general meaning of the word, prices are “the terms on which alternatives are offered” (Wickstead). Thus the wrong-headed concept of “priorities” - a contradiction in terms, the ultimate one-word oxymoron – may be forgiven to non-economists in common speech but is unforgivable when voiced by economists or people in charge of the economy talking about economics.

When there is more than one declared priority, nothing has priority. The Soviet economy was organized - to a different extent at different times in its history - mostly according to a system of multiple priorities, instead of a system of prices: the result was necessarily disorganization and chaos, ultimately leading to economic collapse. Multiple priorities were at their peak during War Communism (1918-21), when towards the end they multiplied beyond belief, to the point that even pen nibs were added to the senseless and useless priority list.

The multiple objectives meaninglessly classed as priorities can be qualitative as well as quantitative, as long as there is a possible conflict and therefore a trade-off between the fulfilment of one qualitative target and another. Indeed what prompted this post is a recent document of the Bruegel Think Tank, on Europe’s economic priorities 2010-2015 – Memos to the new Commission, Edited by Andr√© Sapir, Brussels, 2009.


Anonymous said...

A rather devastating critique. But suppose we rank objectives from the viewpoint of the elasticity of their degree of fulfilment with respect to the application of incremental amounts of money, time, or effort. Would this still come up against your critique?

D. Mario Nuti said...

Your procedure is ingenuous, but is still liable to a small and to a more serious objection.

An objective might only increase its fulfilment by steps, through non marginal applications of money/time/effort, which complicates things no end.

More to the point, although elasticities are dimension-free (so much % increase in an objective's fulfilment for a 1% increase in the application of money/time/effort), the problem remains of taking into account the relative weight attached by the decision-maker to a given % increase in the fulfilment of different objectives.

Whoever used the procedure you suggest would simply replace the decision-maker's valuation of different objectives with his/her own valuation, quite arbitrary and debatable and ultimately irrelevant.

Anonymous said...

Of course you have a point. But can you not make a simple mental adjustment, and understand "preferences" whenever anybody talks about their "priorities"? This is all they mean.

D. Mario Nuti said...

I can assure you that caviar and champagne figure very prominently in my preference system. Yet in my ranked "to do" list - which is what I take "priorities" to mean - the purchase of caviar does not figure at all, because it is doubly expensive both for my money budget and my cholesterol budget; champagne figures only occasionally, for New Year's Eve, anniversaries and special celebrations.

The point is precisely that preferences are only half of the "priorities" problem, the other half is the terms on which alternative objectives can be traded off.

The kind of preferences that could be expressed as a ranked list are rather special. They are called lexicographic preferences, whereby the decision maker infinitely prefers one good (X) to another (Y). Among available alternatives, the one offering most X will be invariably chosen regardless of the amount of Y on offer. The amount of Y will be considered only at equal amounts of X. Again, the lack of a trade-off is a precondition for ranking alternatives absolutely.

Besides, "preferences" may well be what people mean when they say "priorities", but if we could presume to know what people really mean when they say something different from what they mean, where would we be?

Anonymous said...

Alright. I would like to re-phrase my comment.

You could look at a set of "priorities" as a shopping list. Surely you have shopping lists if not a set of "priorities"? Politicians tell us what is on their shopping list, they commit themselves to "buy" more or less of each item on the shopping list as opportunities arise, depending on availability and price in the supermarket where the list is checked out.
What is the matter with that?

D. Mario Nuti said...

Your re-phrased comment is similar to that of my philosopher friend, reported and answered in my original post. My objection to her – that ranking of objectives must depend on their trade-offs, on what we could call their “opportunity cost” relative to the decision-maker's preferences – holds also for your comment.

Shopping is a routine, habitual activity, weekly or even daily. When I draw my shopping list I already know how much I am prepared to spend and the absolute and relative prices of the goods on my list, unlikely to vary significantly and unpredictably from one day to the next. As I know the “opportunity cost” of each item, I can draw a shopping list. But the list will not be ranked, or “prioritized”. The question of prioritizing a shopping list arises only if and when there is an unespected change, for instance when inflation makes the items on my habitual shopping list unaffordable, in which case my objection to “priorities” again holds.

A political manifesto drafted every five years presumably does not contain routine measures, it will be an agenda for change. Remember? “Brown promises: new Government, new priorities”, and it wasn’t even a new government party, for he took over from Tony Blair unelected as premier, though unopposed. An agenda for change cannot be likened to a shopping list.

Besides, if you interpret a government’s “priorities” as a shopping list you will have debased the concept of “priorities” to such an extent that I do not see much difference left between my criticism of “priorities” and your approach.

Anonymous said...

When you go to the supermarket and find that unexpected inflation prevents you from buying all the goods on your shopping list, will you not cancel the item/items with a lower priority?

D. Mario Nuti said...

Normally, if confronted with unexpected inflation, I would not cancel any item on my shopping list, but I would reduce the quantities purchased of each item. In order to do so I would need to know the goods prices and to consider what value I would get from each good whose consumption I cut, relatively to its price. Therefore priorities would not come into play at all.

With two qualifications. First, some goods are indivisible: I can buy a bottle of wine but not a fraction of a bottle (even though bottles come in different sizes. So if my desired consumption of wine falls below one bottle I will have to adjust the time pattern of my purchases and buy a bottle every now and then. So, I may cancel wine from my shopping list occasionally, without signifying a lower priority for wine.

Second, I may have an absolute preference for a bunch of goods in my shopping list, and consider a less desirable good only after reaching a certain amount of that bunch of goods. This is the case already mentioned in an earlier comment, of lexicographic preferences, that do indeed lend themselves to the concept of priorities.

hatfield girl said...

"The language of priorities is the religion of socialism"
as Aneurin Bevan declared at the Labour Party Conference in Blackpool in 1949. But then he would wouldn't he.

D. Mario Nuti said...

This is a wonderful quote, thank you. It forcefully exposes the muddled thinking, the arrogance and sheer impossibility of the social-democratic project.
Yes, he would. And we can all see the utterly disastrous results of that approach.