Available now: USAvailable now: UK

Black Turks, White Turks  

In our last post, we briefly discussed the early history of modern Turkey. If, as we argued, there are indeed powerful continuities between Ottoman institutions and those of the early Turkish Republic, then the continuities between the early years of the republic and today shouldn’t surprise anybody. 

In the interim, Turkey swept aside one-party rule and had periods of fairly competitive democratic elections. But this has always been in the shadow of three social processes inherited from the early republic: a powerful military, weak state institutions with little independence from those controlling the government, and deep divisions within society.

One aspect of these divisions, which has been particularly defining for recent history of Turkey, was pithily captured by Prime Minister Tayyip Erdoğan when he said (see here): 

In this country there is a segregation of Black Turks and White Turks. Your brother Tayyip belongs to the Black Turks. 

White Turks here are the well-educated, well-to-do Kemalist elites fashioning themselves on (some of) Atatürk’s ideas. They are often associated with state bureaucracy and the military. Black Turks are those that the White Turks see as low-educated, lower-class and either still peasants or unable to have shaken off their peasant heritage. 

Even if the terminology of Black and White Turks entered the Turkish lexicon only recently, to understand what Tayyip Erdoğan meant we should go back to the divisions and fault lines that had already become apparent in the Ottoman society in the 19th century. 

Religion — Islam — played a central role in Ottoman society, which comprised mostly of agricultural workers, with limited education. As we discussed in Why Nations Fail, the central state went to pains to limit dissemination of information as part of its control of society. Though for several centuries the central state was able to dominate religious practice and thought, including the religious scholars, the ulema, religious belief inevitably has its own dynamics. These dynamics became stronger during the 19th century as the previously subservient non-Muslim populations rose up and often successfully gained their independence in the European territories of the empire. 

The 1909 ‘counter-revolution’ against the CUP (the Committee of Union and Progress) regime we discussed in the previous post had its roots in the reaction of the lower level ulema, even if it probably was engineered and funded by Abdülhamit. (It is interesting that, witnessing the subservience of state institutions and even the religious-legal establishment, the upper echelons of the ulema appear to have been happy to go along with the CUP, which now controlled the government.) They were able to tap into conservative attitudes, aroused because of perceived threats to their way of life and religion, and the established order.

The leadership of Atatürk’s CHP (Cumhuriyet Halk Partisi, or The Republican People’s Party), after initial attempts and failures to mobilize the religious conservative attitudes of much of the population, started viewing religion and large part of the peasant majority as obstacles to their project. No wonder then that many of the reforms of the CHP were directly targeted at repressing religion, silencing and sidelining this majority, and imposing a new language, historiography and culture on the Turkish people. 

The military and political dominance of the CHP, and their willingness to use robust force if necessary, ensured that the Kemalist project succeeded until the end of World War II under the auspices of the one-party (CHP) rule. Cracks in the system had already appeared, however. In 1946, the Democratic Party (DP or Demokrat Parti) was founded by previous members of the CHP. Though the CHP leadership initially thought that they could control this new party, this proved impossible. In 1950, the DP swept to power with a landslide election victory. Their deputies, and certainly their supporters, were more rural and had a stronger background in commerce and small and medium-size business (contrasting with the bureaucratic or military background as the majority of CHP deputies had). This was the beginning of the political ascendancy of what would become known as the Black Turks. 

We will discuss the economic implications of the DP regime in a later post. For our purposes here, the end of the DP period is more relevant. On May 27, 1960, the first of many military coups in Turkey took place, putting an end to the first Turkish experiment with democracy. The military swiftly moved to hang three on the leaders of the DP, including the prime minister, Adnan Menderes. 

The DP and Menderes had certainly turned authoritarian toward the end of their rule, using repression and heavy-handed censorship of the media. But one can also see the roots of the military coup in the DP’s ability to mobilize large segments of the population until then excluded from politics and also its success in utilizing religion for political purposes. Strikingly, the military coup against the ever first popularly-elected government of Turkish history received the enthusiastic support of the ‘Turkish left’ because the Turkish left had its origins in state bureaucracy and the military and thus identified with the CHP — and hence just like in Latin America, making a mockery of the right vs. left distinction in Turkey. 

The conflicts within Turkish society that came to the surface and then to the boiling point in the next several decades (which we will discuss in our next post) weren’t just a repeat of those between the CHP and DP elites and their supporters. Rather, the brief democratic episode of the 1950s proved that elections wouldn’t easily bring inclusive politics to Turkey. The CHP elites could not stomach losing the election, so immediately started to work to undermine the DP government. The DP elites in return had no compunction in using their power and control over the state, temporary though it proved to be, to undermine the CHP (even attempting to close the CHP in the same way that Kemalist forces had previously and would subsequently try to close opposing parties). This pattern, made more likely by the lack of independent state bureaucracy and institutions, would repeat throughout the next 50 years. 

Indeed, the parallels between the AKP and the DP are hard to miss. So it should be no surprise that the battle lines were also similar, and the conflict that pitted the Kemalist forces of the CHP and military elites against the ‘religious populism’ of the DP would reemerge in the conflicts between the new Kemalists on the one hand, and the AKP and the Black Turks on the other. 

In the intervening years, the military came to increasingly define itself as the defender of all parts of the Kemalist ideology. It purged officers that it suspected of strong religious beliefs or insufficient Kemalist zeal. 

Then, perhaps predictably, the lack of trust and spiral of animosity between the Kemalist elites and the DP of the 1950s reached alarming proportions in the 2000s as the Kemalist political establishment and especially the military immediately took a hostile attitude towards the AKP (whose leader, Tayyip Erdoğan, was barred from taking part in the 2002 elections and could assume power only later after being cleared from the charge of inciting religious hatred). 

The military swiftly moved to threatening another coup against the AKP with a memorandum on its website in April 2007 after the AKP gained control of the presidency (the military had taken actions three more times against elected governments between 1960 and 2002, including in 1997 against the Refah Partisi which could be seen as a predecessor of the AKP even though AKP has partly distanced itself from this party and has generally taken more liberal attitudes on a range of issues). It also threatened action against the AKP from the get-go. Ominously, the Constitutional Court started proceedings to close the AKP because their religious outlook (‘plans’?) violated the constitution. Particularly emotive was the fact that the wife of the new president, the number two of the AKP, Abdullah Gül, wore the headscarf, something banned in public spaces by the Turkish Constitution. 

But the situation was different in 2007 than in 1960. The AKP had already organized deeper social networks within modern Turkish society, and had taken control of large parts of the bureaucracy and the increasingly heavily militarized police, while the status of the military within Turkish society was at an all-time low. This time Kemalists and the White Turks lost.


Institutional Continuities  

Much of the historiography of modern Turkey emphasizes the major reforms and revolutions leading to the founding of the Turkish Republic. One might then have expected a rupture from the Ottoman institutions we discussed in the previous post. But the importance of modernizing efforts, coups and revolutions during this period of modern Turkish history notwithstanding, there are also powerful continuities between Ottoman and modern Turkish institutions.

In fact, one aspect of institutions has remained in essence largely unchanged, casting a long shadow on modern Turkish society: the absence of state institutions and bureaucracy independent from the ruling elite.

As we have seen, state institutions and bureaucracy (including the judiciary) were under the direct command of the Sultan and the ruling elite and thus had little independence during the centuries of Ottoman rule. This changed little after the modernizing reforms started in the 19th century, or after so-called Young Turks, especially the powerful İttahat ve Terakki Partisi, the Committee of Union and Progress (CUP), ascended to power, or after the republican period, in which a new group of officers who had cut their teeth with the CUP started to dominate politics.

A central cause of this continuity lies in the motivation and nature of the institutional reforms in the Ottoman period. These were not implemented, or even supported, by a broad coalition nor were they an indirect response to the pressures and demands from non-elites in society. Rather, they were the outcome of an effort by a narrow and relatively privileged segment of society to strengthen the state capacity and modernize the institutions and the economy in part in response to the financial, economic and military crises facing the empire.

This is not to deny the importance of some of the reforms nor the presence of some more ‘radical’ visions of institutional transformation. For example, though in part a concession to European powers, particularly Britain, in exchange for support for the Ottoman struggle against Mehmet Ali who had declared the independence of Egypt from Ottoman rule, the reforms promulgated with the Gülhane (Rose Garden) edict, which started the era of the Tanzimat (meaning ‘reforms’) on November 3, 1839, had many truly revolutionary elements. Perhaps most radical were the declarations of equality before the law (which were until then essentially unthinkable and unfortunately didn’t really mean much thereafter) and “guarantees for life, honor and property of the Sultan’s subjects” which in practice meant guarantees of the rights of these subjects against the Sultan, who until then did not recognize such rights.

But much else in these reforms was part of a project of modernizing the empire with military, internal administrative, legal and financial reforms to strengthen the empire, without a fundamental transformation of political and economic institutions. This was witnessed by the fact that many of these reforms had no popular basis (nor would it have occurred to any of the leading figures to seek popular support).

The first parliamentary elections under a new constitution were held in December 1876 and January 1877, undoubtedly a landmark event, even if it took further military defeats in the Balkans and the deposing of two monarchs to bring the elites to this point. But this parliamentary period was cut short when Sultan Abdülhamit II closed down the parliament indefinitely and started a period of absolutist rule and repression.

Opposition to Abdülhamit was spearheaded by the Young Turks, a group of young officers and bureaucrats (trained in the newly-founded civil service and war academies), particularly those who had organized themselves as the CUP, founded in Paris and operating in Ottoman lands as a clandestine organization. Though members of the CUP opposed Abdülhamit and wished to reform the Ottoman economy, society and military, their aims were to preserve and strengthen the Ottoman state and take control of it themselves (in this, they were similar to the Meiji reformers in 19th-century Japan, which we discussed in Why Nations Fail, though they never attempted or articulated as radical an institutional transformation).

The influence of the CUP in Ottoman politics was constant from the 1890s onwards, even if their involvement in the government ebbed and flowed. Most important for our emphasis here is that the CUP never envisaged the creation of more inclusive political institutions. Nor did it attempt any form of grass-roots organization. All of this was made quite apparent after the revolution of July 1908, led by CUP officers from the Macedonian and Thracian armies, which ended Abdülhamit’s absolutist rule, bringing back a constitutionalist regime with many of the leading figures of the CUP in charge (even though the CUP itself then and thereafter remained a secret organization). They immediately turned against anything they deemed as opposition. For example, when the political demonstrations that had started against Abdülhamit’s regime continued and turned into strikes and demands for wage rises by workers, the CUP-dominated government reacted strongly, repressing demonstrations and banning strike activity and trade unions. They were interested not in building a regime with broad support, but in monopolizing political power themselves.

The CUP monopoly of power became more naked after 1909, in response to the ‘counter-revolution’ against their control of government (most likely supported and funded by Abdülhamit). In the subsequent years, the government was controlled by the army and a cadre of military officers, but with the strings in the hands of leading CUP figures, particularly after 1913, now intent on consolidating their power by curbing all sorts of freedoms and suppressing any dissent.

There had been previously more liberal elements in the reform movement including Prince Sabahattin (interestingly, long seen as a reactionary figure in republican Turkish historiography because of his liberal, rather than statist, economic views). But they were quickly sidelined by the CUP.

The CUP dominance and repression increased in the years leading up to the World War I, particularly after their coup in January 1913 (following their sidelining earlier in the previous year). This was the prelude to the Ottoman entry into World War I engineered by the CUP leadership and the later massacre of hundreds of thousands of Armenians most likely planned and orchestrated by the same leading CUP figures.

At some level, that the CUP sought to build not any form of inclusive institutions but its own dominance and repression shouldn’t be surprising. It was an organization coming from the military, with no clear social or economic basis and certainly not representing a broad coalition within Ottoman society that would have benefited from more inclusive institutions (and it is questionable whether any such broad coalition could have formed in the Ottoman society). They were swept to power not by opposing the strong power of the Ottoman central state, but because of the weakness of the Sultan and because they articulated a strategy for further strengthening the Ottoman state domestically and internationally — and thanks, in no small part, to their ability to wield military power. Once in power, they were not under pressure to introduce constraints on the exercise of the power of the Ottoman state nor did they have any reason to do so.

How the CUP engineered the Ottoman entry into World War I gives an illustration of how the upper cadre of the CUP wanted to — and, given the concentration of quasi-absolute powers in their hands, could — conduct state affairs in essentially the exact parallel of the Ottoman Sultans’ absolutism. A handful of Young Turks, including two of the de facto leaders of the CUP, Enver and Talât, secretly negotiated a treaty with the Germans one day after Russia had declared war against Germany, committing the Ottoman Empire to also join the war.

The Turkish war of independence in the aftermath of the Ottoman defeat in the war and occupation of several parts of the empire by the Allied powers was spearheaded by another group of young officers, with Mustafa Kemal (later to be called Atatürk, ‘the great Turk’) at the helm. Their approach and attitudes were very much shaped by the CUP, of which many were members.

The Turkish Republic, founded in 1923 following the victory in the war of independence, was undoubtedly a more radical departure from many Ottoman institutions, particularly in abolishing the monarchy, modernizing state bureaucracy, regulating the role of religion, partially liberating and empowering women, and its intention to industrialize Turkey. But one aspect of the Ottoman institutions was never challenged: now state institutions and bureaucracy would be under the command of the new ruling party, Cumhuriyet Halk Partisi, CHP or the Republican People’s Party. Once again, there was no need for broad-based support. In fact, the reforms were formulated with the intent of being imposed on the population that was presumed, rightly, to be opposed to several aspects of them.

With this background and intentions, it should be no surprise that the new republican leadership also did not feel compelled — or wish — to develop the independence of state institutions or other institutional constraints on the exercise of economic, political and social power.

The one-party regime with the Atatürk’s CHP ruling supreme could not of course last forever. It came to an end in the aftermath of World War II, but the institutional legacy tracing back to the Ottoman Sultan’s absolutist control of all state institutions would not only last until then but would also condition the path of democratic politics thereafter, as we will discuss next.


Ottoman Heritage  

In our previous post, we wondered how, after a propitious start, moves towards greater democracy and civil rights in Turkey have stopped and perhaps even gone in reverse.

It will be no surprise to readers of Why Nations Fail that we see the roots of the current problems in the history of Turkish institutions, which trace back to the Ottoman heritage.

We argued in Why Nations Fail that the Ottoman Empire was a perfect specimen of absolutist political institutions, concentrating power in the hands of the Sultan and a narrow elite. First and foremost, this was the result of the highly militaristic nature of the Ottoman state, organized around conquest and expansion, with the role of the Sultan mainly as the commander-in-chief.

The society was explicitly divided into two classes. The ruling class consisted of the military and the state bureaucracy. The rest, referred to as reaya, literally meaning “the flock”, consisted of everybody else doing all the economic activities, mostly centered on agriculture. The military came in part a quasi-feudal arrangement called, the tımar system, whereby lands owned by the Sultan were leased out in exchange for a number of soldiers to be made available at the time of war. From the 14th and 15th centuries on, the military, and to some degree state bureaucracy, came to rely increasingly on the devşirme system, which involved children being taken from the non-Muslim (Christian) subjects of the empire, converted to Islam, and brought up without any political allegiance other than to the Sultan. In fact, the system was largely the outcome of a strategy by the central state to reduce its dependence on tımar holders, who might have otherwise have developed the economic and military power to challenge the Sultan and his bureaucracy. Other strategies involved moving tımar holders large distances within the empire, taking their lands away periodically, and granting tımar rights to soldiers and bureaucrats from the devşirme system in order to prevent the formation of local alliances and concentration of power to rival the central state. The increasing proportion of grand viziers coming from the devşirme system, rather than from local notables and established families, was also part of this strategy. 

Not only military but other political and social institutions were structured so as to minimize constraints on the Sultan and the central state’s power. State bureaucracy, including what we would today view as the equivalent of the judiciary branch, was an instrument of imperial power and thus fully subservient to the Sultan, who was effectively the law and thus above the law (both public and criminal law were based on the Sultan’s decrees). This arrangement naturally removed other potential institutional constraints on the ruler’s power. It was only much later in the history of the Ottoman Empire that parts of the military became sufficiently empowered to resist attempted reforms, and this was only in the context of military reforms targeted at their own removal. 

An interesting argument by Noah Feldman, in The Fall and Rise of the Islamic State claims that the ulema, the religious-legal establishment of Islamic societies, acted as a constraint on rulers. Whether this is true in other Islamic societies is debatable, but it is far from the case in the Ottoman Empire, where the ulema were also integrated into the state bureaucracy under a hierarchical structure under the control of the Sultan and the ruling elite. In fact, the Sultan was in many ways the most powerful representative of religious power, even before but particularly after the caliphate passed to the Ottomans (one of the Sultan’s titles was “the shadow of God”). Defending the Islamic faith and community against the outside world and particularly the infidels was the Sultan’s charge.

Limited private property rights in land (in many core parts of the empire only in less than 10% of the land), the successful strategies of the central state in preventing the emergence of economically and locally independent, powerful tımar holders, the dependence of merchants on the state for many of their activities, and the non-autonomous political and economic status of the cities also implied that economic constraints on royal power which sprang up in Western Europe also failed to develop in the Ottoman Empire.

Other aspects of social institutions were also highly hierarchical — for example, completely eschewing the notion of a single law applying to all subjects — thus creating an environment more conducive to the survival of extractive, absolutist institutions.

All of this is not uncontroversial, however.

First, parts of it sound similar to the thesis of ‘Oriental despotism’ going back to Montesquieu, Hegel, Marx (in the form of the ‘Asiatic mode of production’), Weber and Wittfogel. But the differences from our account are noteworthy. Those thinkers saw something qualitatively different in Oriental societies. We see them as attempting the building of extractive political institutions under their control in exactly the same way that European monarchs also did (and in some cases such as the French almost succeeded). In fact, there is a huge literature, mostly by Turkish scholars, criticizing the view that the Ottoman Empire was a highly absolutist state, an example of Oriental despotism. Some of this is on target as a criticism of the ‘eurocentric’ Oriental despotism school, but does not contradict the strong absolutism of the Ottoman state and the almost complete absence of institutions constraining the power of the ruler and ruling elite (and some of it is very very off target…).

Second, even the most absolutist states of the time, given resources and technology, were much less able than the modern nation-state to regulate economic and social activity over its territories. In the Ottoman case, many parts of the empire, particularly much of the Middle East and North Africa, were peripheral and the state was not even present to keep law and order — let alone restrain extraction by local elites.

Third, extractive institutions in general, and strongly absolutist ones in particular, are neither badly designed nor static. In fact, only great organizational skills and almost farsighted institutional design on the part of the early Ottoman rulers could have turned a little insignificant polity on marginal Anatolian lands into a worldwide empire to be reckoned with. Nor were these institutions static in any sense. They changed and adapted to circumstances after the Ottomans started expanding, in some sense, perfecting their absolutism during this phase. They changed and adapted again when Ottoman expansion came to an end (as shown convincingly by Şevket Pamuk in “Institutional Change and the Longevity of the Ottoman Empire, 1500-1800”.

This institutional heritage is very consequential. We’ll argue in our next post that several aspects of these institutions survived the collapse of the Ottoman Empire and formed an important part of the institutions of the Turkish Republic.


What's Afoot in Turkey?  

In mid-2000s, Turkey was becoming fast the darling of the international community. The economy was booming. Turkish democracy seemed more secure, even vibrant, for first time in decades, perhaps ever. After all, a party standing very much against the principles of the meddlesome and the all-powerful Turkish military, the Justice and Development Party (known by its Turkish acronyms, AKP, for Adalet ve Kalkınma Partisi), had just catapulted to power, winning 34% of the national vote in the November 2002 elections. The AKP — ‘mildly Islamist’ or with ‘roots in political Islam’ as it is sometimes referred to in foreign media — even withstood much bullying and intimidation from the staunchly secularist military, which threatened a coup with a memorandum on its website in April 2007 as the AKP was gaining control of the presidency. Other parts of the anti-democratic Turkish establishment were also gunning against the AKP; the Constitutional Court, for example, attempted to close the party. But the AKP, and in the process Turkish democracy, survived.

What’s more, a peaceful solution to the Kurdish problem appeared on the horizon (the official line was shifting from the denial of the existence of Kurds, viewed simply as ‘mountain Turks’, to allowing Kurdish language broadcasts and publications and even some degree of local autonomy). And Turkey was taking surefooted steps towards accession to the European Union, a process promising to anchor democracy, human rights and civil liberties to European standards.

Then it all capsized. 

To be sure, the love affair of the foreign media with the AKP and its charismatic leader, Recep Tayyip Erdoğan, did not end all at once. As late as last year, the favorite Western commentary on Egypt was to suggest that they should emulate Turkey (and perhaps they are!).

The Turkish economy is still growing though more slowly, and the current account deficit has now reached an alarming size, make any sudden stop a real danger (and the bubbly housing market might still turn south).

More importantly, it’s becoming harder for even outsiders not to see the problems in Turkish politics.

A few snippets should give a fairly good sense of how Turkey is fast turning from democratic haven in the Middle East to unblemished authoritarianism.


  • The AKP is still in power and now with a more solid majority than in 2002, but has turned increasingly authoritarian. The Freedom House, a reliable source on civil and political freedoms and democratic rights around the world, has just downgraded Turkey’s record on civil liberties. Turkey is now in the same boat as Liberia, Morocco, Nepal and Venezuela as far as its respect for civil liberties goes.

  • Turkey has surpassed China in the number of jailed journalists according to The Committee to Protect Journalists, which states that Turkey’s press freedom situation has reached a crisis point, and reports that currently 76 journalists are in jail. The Turkish Journalists Union puts it at 98, far exceeding the number of journalists in jail in the paragons of press repression such as China and Iran (but before one gets nostalgic about the good old days, The Committee to Protect Journalists reminds us that even more Turkish journalists were in jail in the 1990s). Beyond jailing journalists (with the brunt falling mostly on Kurdish ones), the government has created an atmosphere of fear among journalists and newspapers, which are now generally quite supportive of the government, particularly after the Doğan newspaper group was cowed into submission with huge tax bills hanging over its head as the sword of Damocles.

  • The Turkish ‘deep state’ is implicated in a wide range of criminal acts, including the murder of the Armenian journalist Hrant Dink. But the Ergenekon trial, targeted at the Turkish deep state and the Balyoz (‘Sledgehammer’) trial, targeted at the military for alleged coup plots, rather than going after the deep state, appear to have degenerated into a witch hunt against an array of AKP opponents, most likely based on manufactured evidence as argued by Dani Rodrik. (Gareth Jenkins provides the best and most balanced account of the Ergenekon case we have seen.)

  • Unsatisfied with its total control of Turkish academia through the Council of Higher Education — a remnant of the military regime that has long worked to extinguish academic autonomy in Turkish universities and has often clamped down on freedom of thought or even on dress styles and head coverings — the government has taken over the Turkish Academy of Sciences and started selecting and appointing members of this body.

  • Authorities have taken an increasingly intolerant stance against dissent, and not only from those in the Kurdish areas (recently, for example, nine lawyers from a human rights group defending victims of state violence were charged with terrorism). In a particularly vicious incident, members of the band, Grup Yorum, were arrested in a left-wing rally, kept in custody for four days and tortured specifically so as to destroy their ability to play music (the lead singer has had her eardrums ruptured and the violinist has her right arm broken). What’s scarier perhaps is that not many people in Turkey seem to care.

So what’s going on in Turkey? And how did we get here from the rosy days of the last decade? 

The next several posts will investigate the origins of Turkish political and economic institutions and the troubled history of Turkish democracy, where we believe some of the answers lie.


The End of Low Hanging Fruit?  

The argument that innovation and technological progress have been slowing down has been making the rounds. It was recently articulated in Robert J. Gordon’s work and in Tyler Cowen’s influential short book, The Great Stagnation: How America Ate All the Low-hanging Food of Modern History, Got Sick, and Will (eventually) Feel Better. It jumped from obscure academic discussion to public debate when The Economist Magazine turned it into a memorable cover.

The idea is simple and has something to it: in the early 20th century there were many — what Tyler Cowen calls — “low hanging fruits” for the world economy to collect such as antibiotics, electricity-powered factories, radio, TV, planes and automobiles, and not least the great innovation featured on the cover of The Economist, indoor plumbing and sanitation. But these have all been exploited. As we run out of low hanging fruit, the argument goes, we are likely to run out of rapid technological progress and growth will slow down.

Two things are absent in this debate, however.

First, much evidence shows that what determines technological innovations isn’t some sort of “exogenous innovation capacity,” but incentives. This point was stated forcefully by the great economist Jacob Schmookler in his Invention and Economic Growth where he argued (p. 206): 

invention is largely an economic activity which, like other economic activities, is pursued for gain…

Schmookler illustrated these ideas vividly with the example of the horseshoe. He documented that there was a very high rate of innovation leading to improvements in the horseshoe throughout the late 19th and early 20th centuries because the increased demand for transport meant increased demand for better and cheaper horseshoes. It didn’t look like there was any sort of limit to the improvements or any evidence of an “exogenous innovation capacity” in this ancient technology, which had been around since 2nd century BC. Then suddenly, innovations came to an end, but this had nothing to do with running out of low hanging fruit. Instead, as Schmookler put it (p. 93), it was because the incentives to innovate in this technology disappeared because

the steam traction engine and, later, internal combustion engine began to displace the horse… 

Plenty of other examples illustrate how technology responds to incentives, and when incentives are there, great innovativeness follows. Amy Finkelstein’s research, “Static and Dynamic Effects of Health Policy: Evidence from the Vaccine Industry”, shows that when the government increases the demand for some vaccines, pharmaceutical companies respond by starting more clinical trials to come up with these vaccines. In a related paper, “Market Size in Innovation: Theory and Evidence from the Pharmaceutical Industry”, Daron and Joshua Linn show that expansions in the market size for different types of drugs driven by demographic changes lead to sizable changes in the discovery of new drugs.

Recent research by Walker Hanlon of UCLA, “The Necessity Is the Mother of Invention: Input Supplies and Directed Technical Change”, provides another telling example from the 19th century. After decades of focus on American cotton and no technological improvements in, among others, Indian cotton, Civil War-era disruption of U.S. cotton supplies to the British industry led to rapid improvements in textile processes using Indian cotton. 

But, because where incentives for innovation will be strong is difficult to forecast, where the new innovations improving efficiency and consumer welfare will come from in the future is also difficult to predict. Who could’ve predicted a decade ago the ecosystem of innovations related to applications surrounding the iPhone and Android platforms? And yes of course, this doesn’t compare to indoor plumbing, but it illustrates both the rapid response of the economy to incentives and the difficulty of forecasting the areas where new innovations will revolutionize.

And this difficulty feeds into a potentially incorrect line of argument: we know the great innovations of the past and they were great, and we don’t know the ones of the future and thus they can’t be great (though it has to be said that the opposite incorrect argument has been voiced probably more often: we don’t know the great innovations of the future but we are sure they are great!).

All in all, the future is not ours to see of course, but it seems like the responsiveness of innovation to incentives so far suggests that so long as incentives continue so should innovation.

This brings us to the second important issue absent from the debate: the main argument of Why Nations Fail is that innovation stems from inclusive institutions. Hence the debate of whether innovation and technological progress will continue to rapidly improve our lives should center on whether inclusive institutions will spread or will retreat. The great innovations of the Roman Empire, for example, didn’t come to an end (and in fact fall into oblivion) because the Romans reached the limits of innovative capacity but because their institutions turned extractive and then collapsed. But institutions have been notable in their absence in this debate, and we would argue they are the key. (For a take on the future of innovation emphasizing institutions, see here).