Links
Translator
Daily Posts
March 2010
S M T W T F S
« Feb    
 123456
78910111213
14151617181920
21222324252627
28293031  
Categories

Strategic Consistency and Managerial Discipline

Strate­gic Consistency

By strate­gic con­sis­tency we mean the organization’s com­mit­ment to imple­ment its cho­sen plan of action.

Most strate­gies are long-​term in nature, but their imple­men­ta­tion gen­er­ally involves sequences of short-​term actions. Suc­cess­ful imple­men­ta­tion of such plans often involves short-​term sac­ri­fices — just like diet­ing or hir­ing a con­sul­tant. (Although we like to think that our enter­tain­ment value alone jus­ti­fies the mod­est fees.)

If short-​term actions are cho­sen expe­di­ently, then there is a strong pos­si­bil­ity that the stated goal will not be achieved. In eco­nom­ics, this is known as myopic behav­ior. In real life, it is impul­sive­ness or impa­tience. (In pol­i­tics, it seems to be stan­dard oper­at­ing pro­ce­dure.) Any num­ber of exam­ples will illus­trate this inef­fec­tive­ness, but we’ll choose two: one famil­ial and one corporate.

Two Exam­ples of Inconsistency

Spoil­ing Chil­dren – most par­ents do not con­sciously plan to spoil their chil­dren. Their goal is to raise inde­pen­dent, well-​adjusted adults.

Of course, fail­ure to do so may be due to dynamic envi­ron­men­tal fac­tors or bad luck, e.g., Robert Frost’s line “…the best laid plans of mice and men go oft awry…,” but we are not inter­ested in this type of fail­ure. We are inter­ested in the inter­nal fail­ure to exe­cute the stated plan to achieve the desired goal.

For many par­ents, it is just eas­ier and less tir­ing to spoil their chil­dren, espe­cially if one desires the chil­dren to be their friends. (Uh, they’re your respon­si­bil­i­ties, not your acquaintances.)

For­tu­nately, we’ve never fallen for their child­ish manip­u­la­tions, and that explains why we live in the grace­ful state known as the absence of text messaging.

Near the time of our old­est child’s birth, we saw research that found that on aver­age infants looked more like their fathers dur­ing the first six months of life than sub­se­quently. The the­ory was that this mask­ing was an evo­lu­tion­ary device to keep the father nearby and increase the chance of the infant’s sur­vival. We were struck by the fact that babies could change their skulls — their very bone struc­tures — to manip­u­late their elders, and we were impressed. We have been vig­i­lant ever since against their attempts at psy­cho­log­i­cal and emo­tional con­trol (and yet have been voted by those in the know as the coolest par­ents in the school; of course, maybe we were just being manipulated).

In terms of goal attain­ment, in the long-​term, cod­dling and indulging won’t make them love you more — or even like you more. In other fam­i­lies we’ve seen it lead to that won­der­ful com­bi­na­tion of depen­dence, resent­ment, and dis­re­spect — almost the exact oppo­site of the ini­tial, intended goal.

Rela­tion­ship Bank­ing – or why buy the cow when the milk is free? In the fifth score of the twen­ti­eth cen­tury — less grandly, dur­ing the 1980’s — many large cor­po­rate banks embraced “rela­tion­ship bank­ing.” The idea was that value could be extracted from cus­tomers over the long-​term “if they only knew us.” In other words, cus­tomers would be more likely to over­pay for ser­vices and loans if they could expe­ri­ence the plea­sure of our com­pany and our com­pany. (Do we need to go any further?)

These beliefs and poli­cies caused many banks to offer cheap loans as loss lead­ers. Unfor­tu­nately, loans and trea­sury man­age­ment ser­vices are not very sim­i­lar to crack cocaine; they are not phys­i­cally addic­tive, and there are many alter­na­tive legal sources.

So what hap­pened? Banks offered loans with below-​market credit spreads so that cus­tomers could get a taste of their ser­vices. Unfor­tu­nately (and pre­dictably), this did not induce cus­tomers to over­pay for the next loan. In fact and instead, it led cus­tomers to expect low rates the next time because of their past expe­ri­ence and the fact that many com­peti­tors were attempt­ing the same loss-​leader strat­egy. The nerve of those cus­tomers act­ing in their own best inter­ests and foil­ing our plans!

Some­how the “long-​term” rela­tion­ship was going to cre­ate value when nearly every sequen­tial, short-​term deal was priced at the mar­ginal cost — at best. Yet, the expla­na­tion for not fol­low­ing the strat­egy (of com­mand­ing pre­mi­ums) was that it was the best way to fol­low it. This was partly due to the fact that lend­ing money isn’t that much dif­fer­ent that sell­ing air­line tick­ets or other com­modi­ties. 1

It was also partly due to the fact that almost all of the lenders were fol­low­ing the same strat­egy. Although this mim­icry isn’t cru­cial in the out­come, it does make it ironic and humor­ous and inevitable! 2

Thus, one could eas­ily imag­ine bor­row­ers chuck­ling as they recall the old rela­tion­ship adage with the same con­no­ta­tions: why buy the cow when the milk is free? 3

Infer­ences: what are they good for?

So, how can one infer an organization’s actual strat­egy and its con­sis­tency with its stated one? Observe the sequence of actions or pat­terns of behav­ior. Some­times short sequences are suf­fi­cient; how­ever, if the orga­ni­za­tion is behav­ing ran­domly, longer sequences — big­ger sam­ple sizes — may be nec­es­sary. 4 As they have the best infor­ma­tion about oper­a­tions, employ­ees often make these infer­ences and notice strate­gic inconsistencies.

Strate­gic con­sis­tency is sim­ply stick­ing to one’s plans, which is eas­ier said than done.

Man­age­r­ial Discipline

Man­age­r­ial dis­ci­pline is tac­ti­cally nec­es­sary for strate­gic con­sis­tency. It involves the effec­tive imple­men­ta­tion of the organization’s plans.

In cen­tral­ized orga­ni­za­tions, it is rel­a­tively straight­for­ward, and usu­ally involves doing things and get­ting things done — per­fect for the MoA (man of action) who enjoys meet­ings, long days, and giv­ing orders. So, from our per­spec­tive given that it straight­for­ward, it is not very interesting.

In decen­tral­ized orga­ni­za­tions, how­ever, dis­ci­pline often involves not doing things and not tak­ing action — not very good for the MoA who wants to boss and “lead.” Yes, it usu­ally involves that dreaded word — dare we write it — think­ing, instead of doing.

In fact, opti­mally con­trol­ling decen­tral­ized orga­ni­za­tions often involves com­mit­ting not to act as well as design­ing schemes to induce oth­ers to act and then mon­i­tor­ing and adjust­ing those schemes to indi­rectly influ­ence sub­or­di­nates rather than the “managing” subordinates directly. That is gen­er­ally less enjoy­able and that is why dis­ci­pline is required.

Unlike most other types of dis­ci­pline, man­age­r­ial dis­ci­pline is espe­cially chal­leng­ing because it requires both self-​control and empa­thy (to under­stand­ing why sub­or­di­nates behave as they cur­rently do and project how they will respond to new or revised controls).

It is can excru­ci­at­ingly painful when it involves let­ting some­one “get away with” some type of dys­func­tional behav­ior, but such dis­ci­pline on the superior’s part is nec­es­sary to pro­vide real auton­omy to sub­or­di­nates. More­over, the dys­func­tional behav­ior is a response to the imple­mented con­trols; so, it flows from the supe­rior anyway.

In our for­mer life as a busi­ness school pro­fes­sor we enjoyed dis­cussing a case where the shirk­ing sand-​bagger was rewarded (for reveal­ing his pri­vate infor­ma­tion) and the hard-​charging, over-​achiever was not because the for­mer behav­ior was, in fact, coöper­a­tive and the latter’s was not.

Many of the MBAs in the class would howl with indig­na­tion over our con­clu­sions. It just didn’t seem…fair, but in the case, coör­di­nat­ing effi­cient shared resource usage and know­ing (and announc­ing) aggre­gate earn­ings esti­mates with cer­tainty were more impor­tant than moti­vat­ing extreme effort in an attempt to, say, increase the mere prob­a­bil­ity of higher earn­ings. Thus, the firm had con­cluded that there was no cheaper way of elic­it­ing the essen­tial and valu­able hid­den information.

The case reminds us of one of our favorite quotes, which we had shared with the stu­dents before hand:

“Lit­tle is gained from becom­ing indig­nant about self-​seeking behav­ior by man­agers. It is only human for man­agers to have their own goals and ambi­tions…. A more pro­duc­tive response is to take as given the manager’s aims, and ask how to design insti­tu­tions that work as well as possible.”

—John McMil­lan
Games, Strate­gies & Managers

In decen­tral­ized orga­ni­za­tions, man­age­r­ial dis­ci­pline requires the con­tin­ual aware­ness and accep­tance of this fact, which is sim­ply the human con­di­tion. It reminds us very much of the dif­fer­ences between orga­ni­za­tions that take then fallen nature of man as a given — though not doom­ing — and those that believe soci­ety could be per­fected if only the imper­fect ones could be elim­i­nated. Take your choice of 20th cen­tury tyran­nies with this work­ing assump­tion, or for the pop-​culturally-​minded, recall the James Bond movie “Moon­raker” from 1979, where only beat­i­ful and smart peo­ple had a right to live (once every­one else was killed).

Copy­right © 2008 Spero Consulting


Foot­notes:


  1. In fact, money is a com­mod­ity, and money prices — inter­est rates — are often mod­eled the same way as com­mod­ity prices. See Black’s Model.
  2. For you econ­o­mists in the crowd, it is a set­ting where a Bertrand model is more descrip­tive than a Cournot Model. With­out col­lu­sion, only two sup­pli­ers are needed to drive prices to purely com­pet­i­tive lev­els.
  3. We under­stand asset specificity/​hold-​up prob­lems and we under­stand the value of long-​term com­mit­ment in cer­tain settings. There is noth­ing spe­cific about one firm’s money ver­sus another’s, and there was no ben­e­fit for cus­tomers to com­mit to be at a lender’s mercy.
  4. Ran­dom behav­ior is not the same as irra­tional behav­ior. Ran­dom behav­ior can be opti­mal. For exam­ple, it may be nec­es­sary to used ran­dom strate­gies to pro­tect pri­vate infor­ma­tion or avoid the pit­falls of pre­dictabil­ity, but that is a story for another day.
Visitor Locations