Available now: USAvailable now: UK

The Egyptian Paradox: Saving Democracy and Setting It Back

The first freely elected government of a country, where a large fraction of society is disenfranchised, disempowered and made to feel like second-class citizens, is ousted, in the name of saving democracy, by a military coup supported by former elites and “liberals”.

And the outcome? Three more military coups and more than 50 years later, a deep chasm in society that is still preventing the emergence of truly inclusive politics.

No, we are not talking about the future of Egypt (not directly in any case).

This is just a description of what happened in Turkey in 1960 as we described previously.

In Turkey, the first transition to a true multi-party democracy took place in 1946 with the founding of the Democratic Party (DP or Demokrat Parti). This was after two experiments with controlled multi-party democracy, where even window-dressing opposition parties formed by confidants of the regime attracted so much support that they had to be closed by Mustafa Kemal Atatürk, the autocratic ruler of the country.

In 1950, to the great disappointment and apprehension of military and state elites, the DP came to power with a landslide election victory, bringing voice for the first time to millions of Turks effectively disenfranchised until then. Inevitably, this involved a populist and Islam-tinged rhetoric.

But DP elites themselves were no angels (is anybody surprised?). Once they saw their popularity slide, they just adopted part of the playbook of their rivals’ wholesale and augmented their huge corruption with repression and a thorough clampdown on the media.

On May 27, 1960, the military engineered a coup, widely supported by the bureaucracy, the intellectual elites, and the supposedly pro-democracy Turkish “liberals”. After all, wasn’t the military just saving democracy from the DP and its populist leader, Adnan Menderes?

The military swiftly moved to hang three of the leaders of the DP, including Prime Minister Menderes.

Emboldened, the military would intervene three more times in Turkish politics in the next 40 years. The roots of the current problems in Turkey partly lie in the polarization that was much deepened by this coup that wrested power from the hands of those who had been made to feel disempowered for so long the first time they held it.

What would have happened without the military coup? Nobody knows. Perhaps

Menderes and other DP elites would irreparably damage the economy or somehow cow into a total submission before the next election to effectively set up their own dictatorship.

Possible. But unlikely. Rather, they would have probably been kicked out of power in the next election, cementing Turkish democracy’s credentials.

Same thing with the Muslim Brotherhood and Mohamed Morsi in Egypt. Sure, the Brotherhood in power had none of the conciliatory, compromise-seeking veneer that it projected before the election, attracting votes from liberals and leftists unwilling to support the candidate of the military and the old regime. Sure, Morsi was turning authoritarian, attempting to bring his people into positions of power within the state bureaucracy (as the military and Mubarak had done previously). Sure, the economy was ailing (though not just because of the Brotherhood’s mismanagement but also because of the natural instability that accompanies such huge social transformations).

What would have happened without the military coup that took place on Wednesday, July 3, 2013, ignominiously kicking Morsi out of power and into military custody?

Again, we don’t know. It is possible that the economy would have been so deeply damaged that even greater and more violent protests would have erupted. More ominously, the Muslim Brotherhood may have taken over the arteries of power so thoroughly that they would be able to set up their own dictatorship, effectively blocking any path that may have temporarily opened to more inclusive political institutions in Egypt.

All possible, but we also would say that these are perhaps risks preferable to bringing back the army with the support of the so-called Egyptian liberals now cheering an effective return to a military-controlled society. In fact, the Rebel movement, which collected over 20 million signatures to call for early presidential elections, suggests that the Muslim Brotherhood could have easily been defeated at the polls, if only its opponents could bide their time.

Just like in Turkey, what Egypt needed was those ascended to power for the first time to peacefully lose an election — not because the other side cannot tolerate the very thought of those that have so far viewed as second-class sitting in the presidential palace but because they just messed up and weren’t governing well. Because they just lost the support of the ordinary people and had to leave the way they came, through the polls.

Just like in Turkey, Egypt needed assurances to both sides that inclusive politics in which every segment of society, regardless of creed, religion, gender and social status, can share power.

Instead, we have in our hands a military coup that confirms the worst fears of a very large fraction of the population — that the so-called liberal elites and the military that have ruled the country for so long will do anything not to share power with them (never mind that Mubarak and his cronies, together with the military, had also effectively sidelined the young and the liberals who have now turned into allies of the soldiers).

How will this segment of society ever trust democratic politics? How can we expect them not to work to undermine their opponents completely the moment they wrestle power nationally or locally? How can we now hope to end the Egyptian iron law of oligarchy?

It looks like, as in Turkey, the path to true democracy in Egypt will be long, arduous and littered with missed opportunities.


Middle Class Rising?  

A popular interpretation of the ongoing protests in Brazil, Turkey and elsewhere, as well as of the Arab Spring, is through the prism of Martin Seymour Lipset’s modernization theory, which sees democracy automatically following prosperity.


There are many variants of the theory, but the one that seems to be the most popular claims that democracy is a luxury good that populations, particularly the middle classes, demand once they are sufficiently prosperous so as not to constantly worry about their survival and meager economic existence. So, the theory goes, pressure for democracy and democratization will arrive only when a prosperous middle class emerges.


This view is not only frequently expressed in the pages of The Economist or The New York Times (see, for example, here, here or here) but is now all over the Turkish media. This perhaps reflects the power, or at the very least the appeal, of modernization theory.


But there is another interpretation. This view also paints the whole affair in a somewhat more benign hue. After all, according to this perspective, Prime Minister Recep Tayyip Erdoğan and his AKP are responsible for the Turkish economic boom and have thus sown the seeds of the current discontent against themselves as an inevitable byproduct of their surefooted stewardship of the economy.


Though the AKP and the prime minister do deserve some credit for the Turkish economic boom of the last 12 years (even if the picture is more complex and nuanced as we have argued here), this view is both incorrect and trivializes the protests.


This is for several reasons. 


First, despite its great intuitive appeal (especially to the well-educated middle classes already inclined to see themselves as the harbinger of all good things including democracy), Lipset’s modernization theory just doesn’t have empirical support.


Though early work by Lipset himself and others, in particular most notably by Robert Barro, did report support for this theory, this was based on cross-sectional regressions, mainly recovering the fact that democracies are more prosperous. 


More recently, our work joint with Simon Johnson and Pierre Yared looked at the relationship between changes in democracy and changes in income per capita (with or without instrumenting for changes in income per capita). The evidence was fairly conclusive: there is no positive impact of income per capita on democracy. This was true focusing on five-yearly, 10-yearly or longer changes, and both in the post-war sample and throughout the 20th century. We found the same thing in a follow-up article when we distinguished between transitions to democracy and transitions to talk or see, and more carefully modeled their co-dependence. 


Of course, this may not be the final word. But the evidence seems fairly clear that if there is going to be any empirical support for a causal impact of prosperity on democracy, it will have to be much more slowly acting, perhaps taking longer than 50 years. We argued that such interactions might exist and reflect not modernization-type forces, but the joint co-evolution of prosperity and democracy starting at critical junctures.


Bottom-line: a 10-year growth spurt will not bring democracy or create huge protests for more democratic politics in and of itself, be it in Brazil or Turkey. 


Second, not only in recent examples, but throughout history, democracy emerges and takes firmer root because of protests and demands from the previously disenfranchised or excluded —-or at least so we argued in our first book, The Economic Origins of Dictatorship and Democracy.


Though the middle class does play a role in the democratization process, it is often not the driver of the protests or even their main catalyst. Democracy arrived in high-growth authoritarian regimes such as South Korea and Taiwan not because of the wishes or the actions of the middle class, but because of the effective protests, in the face of repression and sometimes violence, organized by students and workers. In Britain, even the landmark First Reform Act of 1832, extending voting rights to the middle class, resulted not because of middle-class protests but because of the Captain Swing Riots organized all over the country by agricultural workers as we suggested in The Economic Origins of Dictatorship and Democracy, and a recent paper by Toke Aidt and Peter Jensen documents. 


Third, the parallels between Brazil and Turkey (and Bulgaria, Indonesia, India and the Arab Spring) should not be exaggerated. Sure we have seen protests in each case. It is also true that in each case they do reflect dashed hopes and disappointments with the government not being as representative of the population at large as it claimed. But fundamentally, they do have different sparks and different origins. They also triggered different reactions. 


In Brazil, they started because of bus fare increases, but more deeply reflect the general dissatisfaction because the social reform process started by Lula and continued by his successor Dilma Rousseff, though it reduced inequality and poverty, did not fulfill the aspirations it set in motion and did not change the fundamentally corrupt nature of politics.


In Turkey they began as protests to stop the destruction of one of the few remaining parks in Istanbul, but quickly turned into protests against the authoritarian style of government of Prime Minister Erdoğan, and perhaps his increasing erosion of certain secular freedoms many Turks cherish.


The reactions have been very different also. President Rousseff, perhaps reflecting her own background as a political activist at the other end of police brutality and torture, or her party’s base of support, has been much more accommodating to protests, even as they turned unruly at times. Prime Minister Erdoğan has taken a much more uncompromising, polarizing hardline stance against the protests.


The consequences are also likely to be different. Though there are already some concessions to protesters in Brazil, the protests are unlikely to change the political landscape there. In Turkey, we argued that they could be a coming-of-age moment for its fledgling democracy.


There are already signs that this might be going on. The mainstream media, which was previously cowed into unwavering support of the ruling AKP, is now filled with stories sympathetic to protesters and opinion pieces on how the prime minister’s hard line has unnecessarily polarize the situation. Perhaps things have already started changing….


(But caution: we also noted that things are likely to get worse before they improve, particularly because the protests and the authorities’ reaction to them will polarize society even further, and this is certainly going on, and will probably continue to intensify for a while yet).


Resource Curse and Institutions: Getting more specific  

In a few blogs over the last several weeks (herehere, here and here), we have argued that the empirical evidence suggests there is a “conditional resource curse” whose existence depends on the institutions of a society. Countries with bad institutions will find that resource abundance lowers their rate of economic growth, while the opposite applies to countries with good institutions.

Yet the bundle of institutions that appear in this literature is very large and encompasses many of the most central political and economic institutions in society. This is important to establish, but it also means that reform of such institutions will be very hard because, as we argue in Why Nations Fail, it is typically not a coincidence that societies have unaccountable political systems, lack the rule of law and have low capacity states all at the same time. In addition, improving the economic consequences of natural resource wealth is probably not the most important reason to reform such institutions now. Nor is it clear that such a focus is the best strategy for doing so.

A good place to start in reforming institutions is perhaps not the macro institutions of the whole society, but the nexus of institutions that surrounds natural resources like oil. After all these resources are owned, mining licenses allocated and the rents distributed in particular ways, all governed by institutions. Wouldn’t these institutions, obviously not picked up by macro measures of the rule of law or checks and balances, play an important role in determining the economic (and political?) consequences of natural resources?

The 2011 book by Pauline Jones-Luong and Erika Weinthal, Oil is Not a Curse: Ownership Structure and Institutions in Soviet Successor States, answers this question in the affirmative.

In 1991 the Soviet Union collapsed and broke up into a number of successor states. In Central Asia this included Azerbaijan, Kazakhstan, the Russian Federation, Turkmenistan, and Uzbekistan. These former Soviet Republics all inherited very similar and weak state institutions, and as they became independent all scored rather badly on the types of institutional measures discussed in the empirical literature on the conditional resource curse.

For example, the Soviet Union had no income tax system for these states to inherit, and many aspects of modern state institutions had to be built up from scratch. In addition to these commonalities of history, all five states were oil rich. Yet very different development paths emerged out of these apparently very similar initial conditions.

While Turkmenistan and Uzbekistan moved onto a classical resource curse type of trajectory, something very different happened in the other three cases. The former two countries expanded the public sector and engaged in grandiose national prestige projects. The latter three actually shrunk the size of the public sector relative to national income.

Jones-Luong and Weinthal argue that this divergence can be traced to the way in which the different countries differed in terms of the ownership structure of oil. They distinguish between four regimes that they argue are critical for determining the consequences of oil.

The first one is when the state owns and controls the oil sector (meaning that they own more than 50% of the shares in the petroleum secto) r. This regime typically involves very limited foreign involvement.

The second is state ownership without control where the proportion of shares owned is less than 50% and where there is more foreign involvement.

The third regime is private domestic ownership of the resource and firms that develop it.

The fourth regime is foreign ownership and control.

There are some simple theoretical relationships between these different regimes of ownership and control and state institutions. Jones-Luong and Weinthal’s main dependent variable is the strength of the fiscal regime and they argue that a weak fiscal regime is a situation where (1) the tax system is unstable, based largely on the natural resource sector and indirect taxation, (2) a system of expenditures which lacks stability and transparency.

A strong fiscal regime is one where the tax system is stable and broad based with a greater reliance on direct taxation and (2) expenditures emphasize budgetary stability and are transparent. The basic theoretical argument is that state elites do not have an incentive to set up strong institutions on their own, they have to be forced into doing so by society.

Different ownership structures of society alter the bargaining relationship between society and the state. For example, if natural resources are privately owned, then the private sector is empowered, and the change in property rights makes it more difficult for the state to just rely on natural resources as its sole tax base. Instead they must develop alternative fiscal resources. Moreover, the private sector can use its greater bargaining power to demand better fiscal institutions since they suffer from bad ones. Hence one would expect that the greater the extent of private ownership and control of resource wealth, the better the fiscal regime would be.

The regimes in the second and fourth categories are intermediate cases and their implications for the fiscal regime are more complex and depend on other factors. For instance, a general feature is that private ownership, by domestic or foreign investors, makes it more difficult for the state to finance itself from resource rents and hence would tend to encourage the development of a stronger stet of fiscal institutions. Yet foreign ownership differs from domestic ownership in that it may be easier for the state to hold up or renegotiate contracts with foreign companies and in the limit, expropriate, foreign companies (as has happened recently in Bolivia and Venezuela and in the past in a whole stream of countries, notably Chile, Iran and Mexico). Thus while one would expect the fourth regime to have better fiscal institutions than the first (state ownership and control), we might expect it on average to have worse fiscal institutions than the third regime.

Returning to the cases, it turns out that indeed Turkmenistan and Uzbekistan had state ownership and control of their oil sectors. Azerbaijan chose state ownership without control, Russia pursued private domestic ownership and finally Kazakhstan chose private foreign ownership. Thus this research suggests that one institution that is critical for certain aspects of the institutional consequences of oil wealth (and by logical implication other natural resource wealth) is the form of property rights and whether or not this is dominated by the state.

Such an argument also probably suggests much more fruitful lines of policy reform than blaming the resource curse on just weak “rule of law” or lack of “checks and balances” —- even though both statements are likely true.


Natural Resources and Political Institutions: The Rentier State

Let us return to the topic of natural resources and institutions.

In a previous post we saw how there has been a debate about whether or not natural resource wealth undermines democracy. The other argument which has connected resource wealth, specifically oil, and political institutions, is that of the “rentier state”.

First made by the Iranian economist Hossein Mahdavy in his 1970 paper, “The Pattern and Problems of Economic Development in Rentier States: The Case of Iran”, the basic thrust of this argument is that if a country is heavily dependent on natural resources, particularly for the financing of the government, then this leads it to have a weak state.

In other words, rentier state = weak state.

It also tends to breed unaccountability since if a state can fund itself via natural resource rents then it does not need to bother developing a fiscal system to tax people. When people are not taxed by the government, the argument goes, they have less incentive to make the government accountable.

These ideas have stimulated a vast literature in social science. They find many resonances in economic history as well. In Why Nations Fail, for example, we discussed how the influx of mineral wealth from its American colonies in the 16th century helped to reduce the checks and balances on the Spanish state but also make it less effective. Indeed, Spain went from being perhaps the dominant power in Europe to an “also-ran” in European geo-political competition.

To our knowledge there is as yet no convincing empirical test of the rentier state thesis possibly because it is very hard to measure state “capacity” or “strength”. 

It seems likely that many of the issues that have arisen in the empirical literatures on natural resources and economic growth and democracy will be relevant.

Maybe natural resource wealth can weaken state capacity in some circumstances, but in others it seems quite likely it could strengthen it. For example, the monarchies of the Persian Gulf had almost no modern state or bureaucratic infrastructure when they discovered and began to exploit oil. Today these states are much larger and more capable even if completely financed by oil. 


Big Data, Big State  

Big data is all the rage nowadays. It is supposed to revolutionize both science and business. The challenge of storing, processing and using the huge amount of information now being made available by our increasingly frequent communications, the Internet, social media and the billions of sensors around the world is surely complex. All the same, we are already generating and using a huge amount of data. The Large Hadron Collider, to pick just one example, is collecting and processing information from 150 million sensors delivering observations roughly about 40 million times per second.

But the state is not to be left behind.

Big data is also whetting the appetite of the state all around the world. With the control of increasing amounts of information, the state can become both more powerful and more intrusive — if it wished and were permitted to do so.

The optimistic perspective would be that if political institutions are sufficiently inclusive, the state will be restrained in its use of information and will only collect, process and utilize the information that it is tasked to by its citizens.

The pessimistic take would be that it is in the genes of the state to control as much and to attempt to become as powerful and intrusive as possible.

Alas, the US evidence so far is more consistent with the pessimistic view.

The recent revelations about the PRISM program of the National Security Agency, showing huge amounts of secret data collection from nine major Internet services and metadata from phone calls show an insatiable appetite for information from the US state and government agencies — ostensibly to stop terrorist attacks, but perhaps for much more.

Looked at from this perspective, another possibility also arises: if this nascent Big State will increasingly try to dominate and control information in the age of Big Data, then it will also tend to take a hard-line attitude against anybody challenging its ability to collect and control this information. If so, perhaps the shock-and-awe attacks by persecutors and agencies against whistle-blowers and rival peddlers of information, such as Julian Assange, Bradley Manning, Aaron Swartz and Edward Snowden, shouldn’t be a surprise.