Daily Archives: August 31, 2012

Valve-like intelligence

You probably remember that a little while ago I gushed over the employee ‘manual’ from the Valve corporation. Following up on that is the attempt, by a Valve employee, to place the company in the context of economic theory and history. Ok, I can’t tell you how much of this is self-congratulatory navel gazing but contained within it is an interesting view of modern corporations.

…there is one last bastion of economic activity that proved remarkably resistant to the triumph of the market: firms, companies and, later, corporations.

Wait, what? Aren’t corporations the engines which drive our capitalist machine? Aren’t they the ‘job creators’? Explain yourself, young man!

…firms can be thought of market-free zones. Within their realm, firms (like societies) allocate scarce resources…Nevertheless they do so by means of some non-price, more ofthen than not hierarchical, mechanism!

Really, it’s quite a simple idea and I’m kicking myself for not thinking about it before like this given all the ranting and raving I do about the inefficiencies and general bone-headedness of hierarchical decision making and resource allocation that I’ve seen. Perhaps I get a point or two for realizing the problem is institutional rather than that of individual decision makers but thinking about it this way does provide a bit of additional insight to the problem. Just because corporations make money – exist to make money – doesn’t change the fact that within their borders they are essentially little Stalinist/Lenninist states with a command economy and centralized decision making. If that system contains inherent contradictions and must inevitably collapse, why don’t corporations?

English: Company logo for Valve Corporation. E... English: Company logo for Valve Corporation. Extracted from http://www.valvesoftware.com Note:Words and short phrases such as names, titles, and slogans; familiar symbols or designs; mere variations of typographic ornamentation, lettering or coloring; mere listing of ingredients or contents; are not subject to copyright. (Photo credit: Wikipedia)

Well, don’t they? Certainly the financial crisis of 2008/2009 looked an aweful lot like the collapsing superstructure of a capitalist system described by Marx (as I understood it) with the exceptions that the state propped up the old system through a huge influx of wealth (mostly coming from…the working classes) AND there still doesn’t seem to be a replacement system waiting in the wings to replace capitalism. And, of course, there’s little reason to think that the troubles are over and that American or European problems won’t continue to hemmorage and accelerate towards collapse.

But, back to the original question. Why haven’t corporations collapsed under the weight of their internal contradictions? I think the answer the author is making is essentially the same one Marx made, because they’re subsidized by the differential between the pay employees get for their labor and the actual value those employees provide. The difference is what Socialists would (I think) call the exploitation of labor. Valve still extracts that extra value from employees to some extent but that revenue stream (at least if I understand what the author is getting at) is limited because Valve itself has to function as a market internally. Employees pick what projects they will work on and can switch if things aren’t working out. This should push down the exploitive component of the firm towards labor and (I suppose) that is recouped at the back end by being able to sell the product for a premium. In short, the internal functioning allows you to resist the constant downward pressure of prices caused by competition by offering a superior product that can demand a higher price.

Ok, so you probably aren’t particularly interested with my growing skepticism of 21st century American capitalism, what does this mean for intelligence analysis?

Right now, the labor market for intelligence analysis is highly inefficient. There are hundreds (perhaps thousands) of government agencies in the United States who employ personnel with an ‘intelligence’ function. The vast majority of those agencies employ a very small number of intelligence personnel (in some cases only one) and moving around between agencies can be very difficult. Within those agencies there are a limited number of types of functions that are conducted. While, in the past, I’ve spoken about the plethora of titles with intelligence personnel can have there are a limited number of skill sets to master, at least within the law enforcement/homeland security fields. The bottom line is that most positions could be filled by people with a broad set of backgrounds supplemented by some reasonable (not onerous) amounts of training.

What might a Valve-like intelligence community look like?

So, imagine a system where intelligence analysts are selected and hired into a large ‘pool’. Perhaps this would be contained within a state or region but, let’s say large enough to encompass one or two hundred analysts and a number of agencies at various levels of government. Each agency that wants analysts would then issue out an initial statement of mission, scope of work, etc. for specific projects and programs and analysts would select which programs they want to work for and they could (at any time) leave one program to work on another that is open.

What? Are you crazy?

No. I don’t think so. Think of the benefits.

You would reduce (perhaps eliminate) the greatest problem faced by intelligence agencies today: lack of focus and purpose. Too frequently, those in charge of intelligence shops are unable to articulate what their intelligence priorities or information needs are. Once you get beyond the insufferable management speak you frequently end up with a blank stare and a fall back to bean counting. If you can’t articulate what a particular program is for then you wouldn’t get any analysts sign up for it. At that point you’d either need to figure out what the hell you’re doing or come to the realization that the program doesn’t have much value after all. Analysts (perhaps like many) don’t like to spend their careers like hamsters on a wheel. They want to do something.

What about programs that are important that can’t attract analysts?

I don’t think that situation would exist. Regardless of the work required, there will be analysts who will want to do it provided it can be demonstrated to make some sort of positive impact. Whether it’s getting bad guys off the street, solving cold cases, identifying future threats in international terrorism or ways vulnerable populations may be exploited, analysts will flock to those jobs. If, instead, the purpose is to garner promotions, count beans or justify grant funding, good luck. Unless you define ‘important’ as something other than the good of the public (like institutional empire building, career progression, etc.) shortages of analysts won’t be your problem. If you can’t articulate why a program is important, it probably isn’t.

Such a program would also ensure analysts are more effeciently paired up with the programs that best suit them. Some analysts love doing very ‘in the weeds’ stuff. Case support, court prep, ‘tactical’ stuff. Others like the more ‘Strategic’ aspects of the job. Still others like a mix of varying degrees. Unfortunately, since intelligence positions are so fragmented and tied to one agency none of that matters. That, in turn, means that you frequently have analysts doing projects that they aren’t particularly interested in. You may want to say ‘Wah, just be happy you’ve got a job you whiner.’ but all else being equal, isn’t it better to have people working on things they have an interest and desire to do? Especially if it’s in the knowledge field? With intelligence work you aren’t sitting at a bench making pins and it’s not a position where you want people to simply go through the motions as if the process was on some sort of checklist. So, keeping people on projects that fit their inclinations and abilities will go a long way to improving performance.

Programs would have lifespans. When a particular program is no longer relevant becaue the threat has become obsolete or a better program is coming into play, analysts will leave and the old program will wither. Institutions can still try to make the case that this or that program is important and relevant but if they can’t make the case in a way that keeps asses in seats they’ll have a tough time when it comes to budget requests.

It would reduce (but by no means eliminate) institutional turf wars. Programs need not be agency specific (although they still could be). The ability of analysts to move between programs would mean that analysts could work with many different analysts and agencies throughout their career with the compositions of those groups shifting all the time. Despite the cliche, I think familiarity here would breed acceptance and cooperation rather than contempt. In part that would be because this system would build a real community among analysts which doesn’t (and I don’t think can) exist the way they’re organized today. Analysts are really creatures of their agencies, even though that results in them often acting in ways that are counter to their professional interests.

So, how do we prevent the analysts from becoming just another interest group that tries to assert power and influence selfishly?

Well, in part by not giving it any structure. There’s no ‘Chief Executive Analyst’ and no ‘Analyst Queen’. Analysts are able to exchange information between themselves and individual decisions will result in benefits to the community. Agency X is run like a tyranny with abusive working conditions? Word will spread throughout the community and that agency won’t get any talent to work there.

Are analysts really the best to decide if a program is worthwhile?

I eagerly await a reasonable alternative. But, snark aside, if a program is worthwhile, someone should be able to articulate why and then, as I say above, getting analysts to cooperate won’t be a problem.

It would allow both top-down and bottom-up projects. While initially I stated that agencies would make proposals and solicit analysts to work them it could work the other way around. An analyst could suggest a particular project for an agency (perhaps a groups of analysts working cooperatively) and agencies (particularly those that don’t have a good grasp of how to integrate intelligence into their operations) could adopt those programs or not. For agencies these could function as pilot projects that would avoid (as we have in the existing system) the cost of implementing a program only to find it doesn’t work and then being stuck with it. For analysts, this would be a positive by (potentially) expanding the numbers of agencies looking to incorporate intelligence into their operations or in the expansion of their use. Greater demand for analysts may lead to increasing the size of the analyst pool and/or increasing the working conditions of analysts (more of that capitalist competition from the beginning of the post).

Does anyone have any questions or see any flaws to this scheme from a practical (rather than political) perspective?