Em­pire The­ory, Part I: Com­pet­it­ive Landscape

[Cross­pos­ted from my Me­dium blog]

Em­pire The­ory is a frame­work for un­der­stand­ing and prac­ti­cing com­pet­it­ive strategy. Com­pet­it­ive strategy is the art of de­feat­ing op­pon­ents. once you have chosen a do­main of com­pet­i­tion, good com­pet­it­ive strategy en­ables you to win.

Com­pet­it­ive strategy re­quires un­der­stand­ing how act­ors be­have based on their po­s­i­tion in a stra­tegic land­scape. This know­ledge serves two clear pur­poses. First, by re­cog­niz­ing the pat­terns of these stra­tegic play­ers, it’s pos­sible to in­fer a vast amount about the stra­tegic land­scape on the basis of re­l­at­ively little evid­ence. Se­cond, a deeper un­der­stand­ing of stra­tegic moves and op­pon­ents’ in­cent­ives al­lows us to bet­ter craft our own com­pet­it­ive strategy, through pre­dict­ing, plan­ning for, and re­spond­ing to be­ha­vior.


Here we use em­pire to mean a group of co­ordin­ated act­ors that op­er­ate around some cent­ral power. Coordin­ated act­ors are those people us­ing dis­cern­ible mech­an­isms for align­ing their ac­tions to achieve par­tic­u­lar goals. A cent­ral power is an actor or set of act­ors caus­ing oth­ers in a given re­gion to co­ordin­ate. The ac­tual cent­ral power may not be the os­tens­ible cent­ral power; for ex­ample, a star­tup might be de facto run by its CTO rather than its CEO. An em­pire then, be­ing a group of co­ordin­ated act­ors, will among those act­ors al­ways have some kind of cent­ral power that is main­tain­ing co­ordin­a­tion. Let’s list some ex­ample em­pires to il­lus­trate:

A com­pany:

Coordin­ated act­ors: Em­ploy­ees, busi­ness part­ners, customers

Cen­t­ral power: The CEO /​ executives

A gov­ern­ment:

Coordin­ated act­ors: The civil ser­vice, the mil­it­ary, cor­por­a­tions, citizens

Cen­t­ral power: the king /​ the pres­id­ent /​ the legislature

The Muski­verse:

Coordin­ated act­ors: People at SpaceX, Tesla, Solar­city, and the Bor­ing Com­pany, per­haps others

Cen­t­ral power: Elon Musk

The Fractal Nature of Empires

Em­pires are fractal. There will be sub-em­pires within any given em­pire. In the Cath­olic church, for ex­ample, we could con­sider the co­ordin­ated act­ors to be the global Cath­olic clergy plus lay people and the cent­ral power to be the lead­ers at the Vat­ican. However, it also makes sense to con­sider a single par­ish as an em­pire where the co­ordin­ated act­ors are the mem­bers of the par­ish and the cent­ral power is the priest. Like­wise, a so­cial move­ment like Ef­fect­ive Al­tru­ism could be con­sidered an em­pire where the co­ordin­ated act­ors are the mem­bers of the move­ment and the cent­ral power is the cluster of people and or­gan­iz­a­tions guid­ing the ideo­logy and strategies of the rest. That said, an in­di­vidual or­gan­iz­a­tion within the move­ment could also be con­sidered an em­pire.

The fractal nature of em­pires fol­lows from the fractal nature of co­ordin­a­tion mech­an­isms. An em­pire can be iden­ti­fied either by no­ti­cing a group co­ordin­at­ing, or by identi­fy­ing a co­ordin­a­tion mech­an­ism and then identi­fy­ing the act­ors co­ordin­ated by that mech­an­ism. As there will be dif­fer­ent co­ordin­a­tion mech­an­isms present at vari­ous places within an em­pire, and thus sub-clusters of tighter co­ordin­a­tion, em­pires will be fractal.

The Con­tents of Empires

Em­pires are com­posed of play­ers, re­sources, and other em­pires. Play­ers are the in­di­vidu­als with enough power to be rel­ev­ant to the over­all func­tion­ing of the em­pire. Re­sources are as­sets that can be drawn upon for the em­pire to func­tion. This cat­egory fea­tures many things be­sides phys­ical re­sources, in­clud­ing money, in­form­a­tion, and per­sonal re­la­tion­ships. Coordin­a­tion mech­an­isms (both nat­ural and con­struc­ted) and people that are not suf­fi­ciently power­ful to be rel­ev­ant for the over­all func­tion­ing of the em­pire are also con­sidered re­sources. Fin­ally, be­cause em­pires are fractal, em­pires con­tain other em­pires.

The Prob­lem of Local Focus

In a given em­pire, the dy­nam­ics of the most cent­ral sub-em­pire have a large ef­fect on the rest of the em­pire, and con­trol of the cent­ral sub-em­pire is im­port­ant to top stra­tegic play­ers as it yields con­trol of the rest of the em­pire. As a res­ult, the top play­ers in an em­pire tend to pri­or­it­ize con­trolling the cent­ral sub-em­pire. This phe­nomenon re­peats in a fractal man­ner. To il­lus­trate, con­sider the Un­ited States an em­pire, and the pres­id­ent of the Un­ited States a player seek­ing to con­trol the em­pire. Within the Un­ited States, let’s say the cent­ral sub-em­pire is the ex­ec­ut­ive branch. Within the ex­ec­ut­ive branch, let’s say the cent­ral sub-em­pire is the cab­inet. If the pres­id­ent can­not con­trol the cab­inet, then it will be much more dif­fi­cult for him to con­trol the ex­ec­ut­ive branch. If he can­not con­trol the ex­ec­ut­ive branch, then it will be much more dif­fi­cult to con­trol the Un­ited States gov­ern­ment.

A great deal of re­sources then tends to be spent on con­trol of the cent­ral sub-em­pire. This al­loc­a­tion of re­sources de­tracts from the proper func­tion­ing of the rest of the em­pire and hurts the em­pire’s ex­pan­sion, as more re­sources spent on cent­ral in­fight­ing means fewer re­sources spent on other things es­sen­tial to the em­pire’s func­tion­ing. Un­for­tu­nately, this out­sized ex­pendit­ure is not the res­ult of cor­rup­tion and whimsy, but polit­ical ne­ces­sity (a lot of what we usu­ally call “cor­rup­tion” stems from polit­ical ne­ces­sity). This prob­lem of local fo­cus is one of strongest lim­it­ing factors on the sizes of em­pires, be­cause the prob­lem tends to get worse as an em­pire gets lar­ger. The prob­lem of local fo­cuses in­creases in lar­ger em­pires be­cause the more power an em­pire has,the more skilled play­ers are at­trac­ted to it. The more skilled play­ers are at­trac­ted to a given em­pire, the more dif­fi­cult it is to con­trol the cent­ral sub-em­pire. The more dif­fi­cult it is to con­trol the cent­ral sub-em­pire, the more dif­fi­cult it is to pre­serve and ex­pand the em­pire. As a res­ult, the prob­lem of local fo­cus hugely lim­its the ex­pan­sion of em­pires.

Power Classes

The co­ordin­ated act­ors in an em­pire will have dif­fer­ing amounts of power. For ex­ample, con­sider a tech star­tup as an em­pire. The founder can hire and fire people, will usu­ally play the lead role in de­term­in­ing the star­tup’s strategy, and can con­trib­ute dir­ectly to the cre­ation of the com­pany’s product. In con­trast, a newly hired pro­gram­mer may only be able to con­trib­ute to the product. As such, the founder has more power in the em­pire than the newly hired em­ployee. Power classes are a ty­po­logy of the co­ordin­ated act­ors in an em­pire on the basis of their re­l­at­ive power levels.

High is the cent­ral power that defines an em­pire’s zone of co­ordin­a­tion. Without high, the em­pire would not ex­ist and the other act­ors would not be co­ordin­ated. High also plays the largest role in de­term­in­ing the dis­tri­bu­tion of re­sources within the em­pire. High can be an in­di­vidual (e.g. a force­ful CEO) or a group (e.g. the board of dir­ect­ors of a found­a­tion). It will of­ten make sense to model high as an em­pire in it­self,be­cause there are nat­ur­ally oc­cur­ring co­ordin­a­tion mech­an­isms that cause high to be its own cluster of co­ordin­a­tion within an em­pire, and there are usu­ally a small num­ber of in­di­vidu­als in high that co­ordin­ate the other high play­ers (a high within high). These nat­ural co­ordin­a­tion mech­an­isms in­clude that high play­ers are mu­tu­ally threatened by middle play­ers and by ag­gress­ive out­side em­pires.

Mid is the col­lec­tion of in­di­vidu­als or groups that have suf­fi­cient power to chal­lenge high’s con­trol. Mid play­ers will of­ten have smal­ler em­pires of their own. Mid plays an im­port­ant role in con­strain­ing the ac­tion of high. In our tech star­tup ex­ample, mid play­ers might be the man­agers of the en­gin­eer­ing and sales teams.

It does not usu­ally make sense to model mid as a single em­pire. They are very sel­domly co­ordin­ated as such.

Be­cause mid play­ers con­trol fewer re­sources than high play­ers, any mid player will have to ex­pend a greater por­tion of their re­sources to se­cure the co­ordin­a­tion of a fel­low mid player. An in­vest­ment of $1M is a not­able and risky ven­ture when your net worth is $20M. It might be an af­ter­thought if your net worth is $2B.

Each in­di­vidual mid player con­trols not­ably fewer re­sources than high, you have to co­ordin­ate more of them to reach the same cap­ab­il­it­ies a single high player can provide. Coordin­a­tion costs are su­per­lin­ear, so pool­ing any­thing ex­cept the simplest re­sources in this way is un­eco­nom­ical. Coordin­at­ing 30 dif­fer­ent stra­tegic play­ers rather than 3, is like­lier to in­crease costs by a factor of 100 rather than 10.

For any given mid player, high is usu­ally a prefer­able ally to other mid play­ers. Given these known prob­lems and the ex­ist­ing un­cer­tainty in mu­tual eval­u­ation, a mid player must then not only match, but out­bid the of­fer high mid give. This is in­fre­quent.

Low is the col­lec­tion of play­ers that can chal­lenge mid but can­not chal­lenge high. Low has the largest pop­u­la­tion and the least power. In our tech star­tup ex­ample, the low play­ers would be in­di­vidual pro­gram­mers or sales people. The pro­gram­mers on an en­gin­eer­ing team could plaus­ibly chal­lenge their man­ager, but they could not plaus­ibly chal­lenge the founder. Like mid, it does not make sense to model low as an em­pire.

Out­side is any actor that is not co­ordin­ated by the high power. In our ex­ample, this could be an em­ployee at an­other com­pany or the mayor of a town in France. Out­side play­ers may still seek to af­fect an em­pire, in­clud­ing by med­dling in its in­ternal af­fairs.

As men­tioned earlier, cer­tain act­ors are best modeled as re­sources. Any actor that can­not in­de­pend­ently chal­lenge mid is best un­der­stood as a re­source, be­cause these act­ors will not be rel­ev­ant for un­der­stand­ing the em­pire. They can be un­der­stood as re­sources, be­cause they will be used by low, mid, and high play­ers to ac­com­plish their ob­ject­ives. For ex­ample, they might provide labor or be weapon­ized by play­ers against each other.

Examples of clas­si­fy­ing by Power Class

In the Un­ited States today, high is best un­der­stood as be­ing com­posed of key fed­eral agen­cies. Heads of ma­jor in­sti­tu­tions such as large com­pan­ies, banks, uni­ver­sit­ies or gov­ernors of in­di­vidual states can be un­der­stood as mid. State of­fi­cials, heads of local groups and smal­ler or­gan­iz­a­tions can be un­der­stood as low. Every­one else is best modeled as a re­source. Rel­ev­ant out­side powers con­sist of key for­eign gov­ern­ments such as Ch­ina or Rus­sia.

At Har­vard University, per­haps high is oc­cu­pied by the pres­id­ent, prov­ost, deans, vice pres­id­ents, or trust­ees. Mid might be key pro­fess­ors, long-time staff, heads of de­part­ments and ma­jor donors. Low might be stu­dent or­gan­izers or less im­port­ant pro­fess­ors. Other stu­dents, as­sist­ant pro­fess­ors, re­place­able staff, and smal­ler donors. Rel­ev­ant out­side play­ers might be com­pan­ies that re­cruit from the uni­ver­sity or the local city gov­ern­ment.

Power Classes are Fractal

Like em­pires, power classes are fractal. The same actor can be clas­si­fied as low, mid, or high de­pend­ing on the frame of ref­er­ence. For ex­ample, a par­ish priest in New York might be low if con­sid­er­ing the en­tire Cath­olic church, mid if con­sid­er­ing the Arch­diocese of New York, and high if con­sid­er­ing the priest’s par­ish it­self.

Cau­tions in Classification

The of­fi­cial story of who is and is not power­ful does not al­ways match the ac­tual story. For ex­ample, it might be that the pres­id­ent of Har­vard has only mod­er­ate in­ternal in­flu­ence and that one of the deans has by far the most in­ternal in­flu­ence. In this case, the pres­id­ent might be bet­ter clas­si­fied as a mid-player. When as­sign­ing in­di­vidu­als and groups to power classes in an em­pire, be skep­tical of your as­sess­ments, as it is easy to as­sume power dis­tri­bu­tions based on the of­fi­cial story.

Stra­tegic Landscapes

A stra­tegic land­scape is a do­main of com­pet­i­tion among play­ers. A do­main of com­pet­i­tion is a re­gion in which play­ers com­pete for scarce re­sources.

Try­ing to ana­lyze a stra­tegic land­scape without spe­cify­ing a do­main of com­pet­i­tion will yield con­fu­sion and er­ror. If the do­main of com­pet­i­tion isn’t spe­cified, ends and means can­not be dis­tin­guished. Most ac­tions are am­bigu­ous, so un­less they are in­ter­preted through a def­in­ite hy­po­thesis, in­vest­ig­a­tion has no clear dir­ec­tion and un­cer­tainty can­not be re­solved.

This ap­proach dis­tin­guishes the mere ac­cu­mu­la­tion of facts from ana­lysis. The cru­cial task is de­term­in­ing which facts are rel­ev­ant and pri­or­it­iz­ing them. While you might ima­gine a lo­gist­ical ana­lysis that doesn’t spe­cify a do­main of com­pet­i­tion, it will fail to pre­dict the range of in­ter­ac­tions between play­ers.

You might cor­rectly note the in­dus­trial ca­pa­cit­ies in a par­tic­u­lar re­gion, but if you are not keep­ing track of whether the factor­ies are aligned either through an owner, a state or an oli­go­poly or whether they are mis­aligned, you will fail to pre­dict which products can be built or which pro­jects will be car­ried out.

Since play­ers can modify any mere lo­gist­ical fact, the ac­cu­mu­la­tion of facts without know­ledge of the do­main of com­pet­i­tion will per­haps cor­rectly show the func­tion­ing of some sys­tems but will fail to pre­dict changes in the sys­tem.

For ex­ample, ana­lyz­ing the stra­tegic land­scape that in­cludes the oil in­dustry and the so­cial justice move­ment without spe­cify­ing a re­source they are com­pet­ing over will res­ult in some­thing like a list of re­ports of me­dia events and gen­eral be­liefs. To un­der­stand their dy­nam­ics or even cor­rectly eval­u­ate the facts on the ground you have to identify either a def­in­ite con­flict point, or their over­all stra­tegic aims and po­s­i­tion.

You might be­gin to ana­lyze them as a part of the polit­ical stra­tegic land­scape in which the re­source com­peted over is the al­le­gi­ance of a par­tic­u­lar con­gress­man.

The oil in­dustry might have the abil­ity to of­fer pos­it­ive re­sources in the form of fin­an­cial or legal sup­port for the con­gress­man. Their pur­pose for com­pet­ing in the polit­ical land­scape be­ing fa­vor­able le­gis­la­tion for their in­dustry. The so­cial justice move­ment might be able to mount a cam­paign against the con­gress­man, at­tack­ing their char­ac­ter. Their pur­pose be­ing so­cial re­form, per­haps through le­gis­la­tion.

In the com­pet­it­ive scen­ario laid out it would only makes sense for an un­pop­u­lar or weak con­gress­man to go with so­cial justice, and only tem­por­ar­ily, since all they can of­fer is to stay their hand, while the oil in­dustry can provide use­ful re­sources, that im­proves the con­gress­man’s long term po­s­i­tion.

Em­pires are do­mains of com­pet­i­tion, and do­mains of com­pet­i­tion tend to be em­pires; em­pires are al­ways do­mains of com­pet­i­tion in which play­ers are com­pet­ing for power, and do­mains of com­pet­i­tion al­most al­ways have co­ordin­at­ing mech­an­isms bind­ing the com­pet­it­ors to­gether (for ex­ample, com­pet­it­ors in the oil in­dustry co­ordin­at­ing to de­feat clean-air le­gis­la­tion).

The term “land­scape” provides a use­ful meta­phor for think­ing about these do­mains of com­pet­i­tion. You can think of the ter­rain of a stra­tegic land­scape as be­ing de­term­ined by the com­pet­it­ors and their re­l­at­ive power. Ima­gine your­self stand­ing on a pre­cip­ice over­look­ing a stra­tegic land­scape of a uni­ver­sity. You see rolling hills off to the left, some of which are lar­ger than oth­ers, rep­res­ent­ing the heads of the vari­ous hu­man­it­ies’ de­part­ments. In the middle is a tower moun­tain rep­res­ent­ing the cent­ral ad­min­is­tra­tion, upon which there is high rocky out­crop rep­res­ent­ing the pres­id­ent of the uni­ver­sity. The land­scape is not static, but dy­namic, with the ter­rain shift­ing as play­ers make moves and gain or lose power. If you want to com­pete in this stra­tegic land­scape, you will have to nav­ig­ate it, tak­ing into ac­count the powers of the other play­ers in de­term­in­ing your path, your com­pet­it­ive strategy. The same goes for the other com­pet­it­ors.


Earlier I claimed that act­ors ex­hibit com­mon pat­terns of be­ha­vior de­pend­ing upon their re­l­at­ive po­s­i­tion in a stra­tegic land­scape. Now we can parse this: in a do­main of com­pet­i­tion, as­pects of the be­ha­vior of high, mid, and low play­ers will be con­sist­ent and re­cog­niz­able. This means, for in­stance, that there are pat­terns of in­ter­ac­tion between high and mid, and that, if we identify high and mid in a par­tic­u­lar do­main, we will im­me­di­ately learn much about how those play­ers will be­have. The com­mon be­ha­vi­ors of play­ers are a con­sequence of what works and does not work for play­ers given their po­s­i­tion on the land­scape. Under­stand­ing such pat­terns thus sub­stan­tially broadens one’s range of avail­able stra­tegic op­tions. We will ex­plore these dy­nam­ics in de­tail in part two.

Contin­ues with Part 2: Power Dynamics