T. Sabri Öncü ([email protected]) is an independent economist based in İstanbul. A slightly edited version of this article first appeared in the Economic and Political Weekly on 12 September 2026.
Image with acknowledgement to Wikimedia Commons – Alfred Friedländer – Looting
To Doruk Madencilik and Eti Gümüş Workers
The Yıldızlar case is not only a story of entrepreneurial failure, but also of successful sabotage. Vested interests have captured the state, used it to collect rents, and obstructed productive alternatives.
Introduction
On 18 July 2022, I wrote an article in this column titled “To Loot or Not to Loot? How Public-Private Partnerships Harmed Turkey.” [1] This article is a follow-up to that article.
In that article, among other things, I argued that Türkiye’s public-private partnership (PPP) experiment was a case study in Akerlof and Romer’s (1993) theory of looting. In this article, moving beyond this PPP experiment, I turn to the specific case of Yıldızlar SSS Holding to extend my earlier analysis, based mainly on Akerlof and Romer, in order to identify a mechanism through which wealth has been extracted in Türkiye since the Justice and Development Party (AKP) came to power in late 2002 (see Öncü, 2022). As I will try to demonstrate, this mechanism relies on interactions between state capture, rent creation and distribution, and the selective use of institutions. Selective use of institutions does not mean that institutions are weak. An institution can be highly active without being effective in the way it claims.
While continuing to use the Akerlof-Romer framework, I bring two additional concepts into my analysis of the Yıldızlar SSS Holding case: Black’s (2005) concept of “control fraud” and Veblen’s (1921) concept of “sabotage”. Recall that Akerlof and Romer’s theory of looting is about how those who control a firm may find it profitable to extract value from it rather than preserve its long-term viability. Black’s control fraud, on the other hand, focuses on the ability of those who control an organisation to subvert the controls designed to constrain them and turn those controls into instruments of fraud.
These two frameworks, however, are not sufficient to fully explain the phenomenon because their primary focus is on the individuals and their ability to subvert the controls around them. But what happens when the institutions that should constrain the firm—regulators, public authorities, courts, or other state institutions—are themselves involved in the fraud? Put differently, what if the state institutions that are supposed to constrain the private actors are themselves part of the rent extraction mechanism? At that point, the problem can no longer be understood solely as the behaviour of an individual or a group of individuals.
Veblen’s concept of sabotage also suffers from a similar problem. For Veblen, sabotage is the deliberate restriction of productive efficiency and capacity for pecuniary purposes. Central to his concern was the conflict between industry and business: for him, industry was the productive organisation of material life, whereas business was for pecuniary gain. From Veblen’s perspective, vested business interests may find it profitable not to increase production but to restrict it, withhold productive capacity, or otherwise interfere with the efficient functioning of the productive system. Again, the same question, albeit stated slightly differently: what happens when the institutions that should constrain the vested business interests—regulators, public authorities, courts, or other state institutions—are themselves involved in the sabotage? Put differently, what if the state institutions that are supposed to constrain the vested business interests are themselves part of the sabotage?
In this article, I begin with Akerlof and Romer’s looting, Black’s control fraud and Veblen’s sabotage “to boldly go where no one has gone before.” [2] That is, I bring these three concepts together and attempt to go beyond them. The concept of the looting state I introduce is to describe a broader institutional configuration and the case of Yıldızlar SSS Holding provides an opportunity to examine this configuration.
Mine is not simply a story of a failing company or an isolated case of corporate fraud or sabotage by vested business interests and the like. It is a story of a broader relationship between private accumulation, state power, rent creation and distribution, selective use of institutions, and the obstruction of productive alternatives.
Let me conclude this introduction by stating that all documents supporting the factual claims made in the rest of this article are available on my personal page to save space, as there are too many to include here.[3]
Yıldızlar before the Holding
Established by Sebahattin Yıldız in 2005, Yıldızlar SSS Holding is wholly owned by members of the Yıldız family. Indeed, Yıldızlar is the plural of the family’s surname. According to its website, the Holding operates principally in mining, ceramics and granite, and energy and petroleum, with additional activities in construction, logistics, insurance and tourism.
The story of Yıldızlar SSS does not begin with the Yıldız family, however. It begins in 1973 with Söğüt Seramik Sanayii A.Ş. (whose logo was “SSS”), a broadly held ceramic venture backed by hundreds of shareholders, with Türkiye İş Bankası (one of Türkiye’s largest deposit banks) as its largest shareholder. By the late 1990s, however, SSS was in serious financial difficulty, not to mention the entire country (see Öncü, 2022; Cömert and Öncü, 2023). The company had financed much of its expansion through bank borrowing, and by 1998 its outstanding bank loans had reached some $60 million, spread across 21 banks. By mid-1999, it could no longer obtain further credit, and a shortage of working capital brought production to a halt. It owed millions of dollars to its workers and various public institutions, and more than 200 enforcement proceedings had been initiated against it.
In 2000, when the company was effectively insolvent, Sebahattin Yıldız, who had previously operated SSS dealerships, established Söğütsen Seramik Sanayi İnşaat Maddeleri İthalat İhracat A.Ş. (“Söğütsen”) and leased the factory for five years at a rent equivalent to 20% of annual production. In 2001, he established 3S Madencilik Seramik İnşaat Nakliyat A.Ş. (“3S Madencilik”) to market and export the ceramics produced by Söğütsen. The same year, Doruk Madencilik ve Elektrik Üretim Sanayi Ticaret A.Ş. (“Doruk Madencilik”) was established with Sebahattin Yıldız as the controlling shareholder, initially as a separate vehicle for mining and electricity generation.
In 2003, Söğütsen acquired the SSS facility through enforcement proceedings, but the sale was subsequently annulled following a lawsuit brought by the Savings Deposit Insurance Fund (TMSF) on behalf of the creditor bank, Pamukbank. Söğütsen nevertheless continued to operate the factory.
This is how it all started.
Before moving on to the next section, let me mention that in 2004, Yaşar Tüzün, a member of parliament from Bilecik Province where the SSS factory was located, stated in parliament that the workers’ claims remained unpaid and that their grievances continued. Also that year, the Capital Markets Board (SPK) found that the leasing relationship between SSS and Söğütsen between 2000 and 2003 had involved concealed profit transfers of about $7 million and ordered the amount, together with statutory interest, to be returned to SSS.
These were early signs of a pattern in which Yıldızlar could leave obligations behind while pursuing their pecuniary interests. The pattern, it seems, has not disappeared: in April 2026, workers at Yıldızlar-owned Doruk Madencilik came to Ankara, the capital city, demanding unpaid wages and compensation, and in August 2026, workers from Doruk Madencilik and Eti Gümüş (Türkiye’s only integrated silver-production company, also owned by Yıldızlar) again came to Ankara over unpaid wages, compensation and other employment rights.
Eti Gümüş Privatisation and the Holding
The privatisation of Eti Gümüş A.Ş. was a major milestone in the history of the Holding, although the Holding did not yet exist. Recall the two Yıldız-controlled firms: 3S Madencilik and Söğütsen. In the first tender, held on 8 May 2003, 3S Madencilik submitted the highest bid, $40 million, but failed to fulfil its contractual obligations and the tender was cancelled. In the second tender, held on 6 May 2004, Söğütsen won the tender with a bid of $41.2 million.
Recall also that Söğütsen was the company that had leased the Söğüt Seramik factory from SSS in 2000 and whose dealings with SSS had later been found by the SPK to involve concealed profit transfers of about $7 million. The financial standing of SSS—which was deeply indebted and effectively bankrupt—was questioned during the privatisation process, and the issue was raised in parliament. The Privatisation Administration, however, said that its records contained no information or documents establishing a relationship between Söğütsen and SSS.
Eti Gümüş was hardly a distressed asset, whose value according to the Turkish Chamber of Metallurgical and Mining Engineers was about $450 million at the time of privatisation. According to figures from the Privatisation Administration, it made approximately $11.3 million in net sales and $520,000 in net profit in 2002. In the first quarter of 2003, net sales were approximately $6.4 million and net profit $330,000. At the time of the sale, the company had 356 employees and more than $20 million in cash and silver inventory, in addition to service buildings valued at about $4 million. Metin Aydoğan (2005) later put the cash and silver inventory at about $20.6 million. The figure is striking: the purchase price was $41.2 million, while the buyer could pay half upfront and the remainder over 24 months. The cash and silver inventory held by Eti Gümüş were therefore roughly equal to the entire initial payment.
Söğütsen did not acquire Eti Gümüş alone. In June 2004, Söğütsen and ATP İnşaat ve Ticaret A.Ş. established KSS Madencilik İnşaat Turizm Sanayi ve Ticaret A.Ş. as the investment vehicle to acquire Eti Gümüş, with Söğütsen holding 51%, ATP 48.7% and three members of the İpek family the remaining 0.3%; the structure was adjusted to 50%-50% shortly after. ATP was 99.04% owned by the İpek family’s İpek Matbaacılık Sanayi ve Ticaret A.Ş. ATP’s and the İpek family’s total contribution to KSS was about $10.3 million. In February 2005, however, ATP and the İpek family sold their 50% interest in KSS to Söğütsen for $12 million, making Söğütsen the sole owner. Why the İpek family was brought into the structure, only to exit for $12 million several months later, remains an open question. A roughly 16.5% return on $10.3 million in such a short period is nevertheless worth noticing, given that such a return was huge in dollar terms at the time.
Shortly after, Yıldızlar SSS Holding was established, Doruk Madencilik was turned into a wholly owned firm of the Yıldız family and the companies were subsequently brought together under the new holding structure. In 2007, Eti Gümüş acquired Ergani Bakır—a major copper facility—from Eti Maden İşletmeleri Genel Müdürlüğü, the state-owned mining enterprise, for approximately $17.5 million.
Thus, the acquisition of Eti Gümüş was not an isolated move into mining but part of a wider corporate structure in which different Yıldız-controlled companies held different mining interests. About the time of this acquisition, the group started accumulating mining licences on a remarkable scale and, according to its website, in 2012, it controlled 2,364 mining licences, around 16% of all mining licences in Türkiye that year. I should mention that while a licence is an economic asset, since accumulating licences can prevent others from developing the resources, it is a form of what Veblen meant by sabotage: not necessarily destroying productive resources, but restricting their use in the pursuit of pecuniary interests.
This is where the story begins to look less like a series of unrelated acquisitions and more like the building of a mining group around former state assets. The looting of a company does not necessarily require the looters to take the company to bankruptcy while enriching themselves. It can also involve acquiring valuable productive assets while leaving obligations elsewhere. And this is where the looting state comes in: if the state allows valuable assets to be transferred while the corresponding obligations are left behind, the state is no longer merely standing by while looting takes place but actively facilitates it.
Osmangazi Electricity Distribution Fiasco
Electricity distribution had ceased to be a national public monopoly by the time Yıldızlar entered the business. In 2004, TEDAŞ—the state-owned Turkish Electricity Distribution Corporation—was reorganised into 21 regional monopolies, which were subsequently privatised. Privatisation therefore did not create competition in distribution but rather transferred the right to collect monopoly rents in each region from the state to private interests.
On 6 November 2009, Eti Gümüş, a Yıldızlar company, won the Osmangazi auction with a bid of $485 million, narrowly beating the next bid. The shares were subsequently transferred from Eti Gümüş to Dedeli Yatırım İnşaat Taahhüt Elektrik Dağıtım A.Ş., a Yıldızlar company established for this purpose, in May 2010, and Akbank—a large Turkish private bank—reportedly financed $375 million of the acquisition. Although the remaining $110 million was described as equity, the underlying financing of that purported equity is not public knowledge.
What was remarkable, however, was that, having acquired one regional monopoly, Yıldızlar was simultaneously trying to acquire seven others. Through Yıldızlar SSS Holding itself and the Yıldızlar-controlled Eti Gümüş–Söğütsen joint venture, it bid for İstanbul Anadolu Yakası, Toroslar, Akdeniz, Boğaziçi, Gediz, Trakya and Dicle regions. At İstanbul Anadolu Yakası, it bid $1.812 billion, losing by only $1 million. At Toroslar, it won the December 2010 auction with $2.075 billion, just $1 million above the next bid. Yıldızlar subsequently failed to complete the acquisition, and Toroslar was eventually re-auctioned and transferred to another company for $1.725 billion in 2013. Even more remarkably, when the winning bidder for İstanbul Anadolu Yakası defaulted, the Privatisation Administration invited Yıldızlar where it was the runner-up, to take over that regional monopoly as well—shortly before Yıldızlar failed to finance either acquisition.
Thus, within little more than a year, Yıldızlar had acquired one regional monopoly and bid for seven more of Türkiye’s 21 electricity distribution regional monopolies.
Where was the money to come from?
That question is difficult to answer because the relevant financing arrangements are not publicly disclosed. The answer that “the rest came from equity” is not particularly informative when the source of that purported equity is itself opaque. “Trade secret” restrictions make it difficult to establish exactly how the acquisition was financed and how the subsequent financial problems accumulated.
A more fundamental question is this:
How was a highly leveraged entity with apparently limited capital allowed to repeatedly bid billions of dollars for regional monopolies?
The answer to what happened next was less ambiguous, though. In 2013, after the Osmangazi companies failed to fulfil their obligations, including a substantial debt to TETAŞ, the state-owned electricity trading company, the Energy Market Regulatory Authority took over their management. The regional monopoly that had been transferred for $485 million had, within a few years, effectively returned to state control.
The Osmangazi fiasco was therefore not simply a story of a private entity that may have overpaid for an asset. It was a case in which a publicly owned regional monopoly was transferred to a highly leveraged private entity and, within a few years, required regulatory intervention when the entity could no longer meet its obligations.
That is the looting state.
From Osmangazi to Adularya
I should mention that Osmangazi was an extraordinary case. It appears to have been the first—and, to my knowledge, the only—privatised electricity-distribution and retail-sales operation whose management was subsequently taken over by the regulator after its private owner was unable to fulfil its obligations, while ownership remained with the private owner. Nevertheless, it shows that privatisation does not end the state’s role in the allocation of assets and rents; it changes the form in which that role is exercised. The state can transfer a public concern to a private entity and intervene when the private structure subsequently fails.
Adularya provides a different example of state intervention. Adularya, a company of Naksan Holding, developed the Yunus Emre thermal power plant in Eskişehir Province, designed to burn lignite from its associated mine nearby, from where Doruk Madencilik workers came in April 2026. The project was financed in substantial part by a €485 million loan from the Czech Export Bank, with Naksan Plastik—a Naksan Holding company—as guarantor. The first unit entered service in February 2016, but the project encountered serious technical and financial problems, and never began sustained commercial electricity production.
Following the failed coup attempt of 15 July 2016, which the Turkish government attributed to the Gülen congregation, many businesses were singled out for state intervention because of their alleged ties to the congregation. TMSF, which had previously been responsible for insuring deposits, and resolving failed banks and other financial institutions, was turned into an institution through which the state could take control of privately owned companies on grounds that were not reducible to their economic performance in the post-July-2016 period. Naksan Holding was among those, and Adularya was transferred to TMSF.
After TMSF took control of the Yunus Emre thermal power plant and the associated lignite mine, the plant remained unable to operate commercially. However, the Czech Export Bank’s €485 million claim against Adularya remained outstanding. In 2021, Yıldızlar SSS Holding approached TMSF to acquire the claim. TMSF approved the assignment in May, and Yıldızlar completed the acquisition in September for reportedly about €70 million: €20 million upfront and the remaining €50 million in 50 monthly instalments of €1 million. Why Czech Export Bank agreed to this deal and how the upfront €20 million was financed are not publicly known because both are a “trade secret”. Only afterwards, in July 2022, did TMSF put Adularya up for auction, which the Yıldızlar company Doruk Madencilik won. The sequence, however, is difficult to regard as an ordinary arm’s-length disposal.
The contrast with Osmangazi is instructive. In both cases, the state intervened after private arrangements failed, but the nature of the interventions was different. At Osmangazi, the state took over management of a previously state-owned entity while leaving ownership with the private owner. At Adularya, it took ownership of an already-private entity, approved the transfer of a major foreign creditor’s claim to a local private entity, and subsequently auctioned the enterprise.
The issue is therefore not simply privatisation, nor simply corporate failure, nor simply control fraud and the like. It is the interaction of state capture, rent creation and distribution, and the selective use of institutions.
That is the looting state.
Ankara from April 2026 onward
In April 2026, more than 110 Doruk Madencilik workers from the neighbouring province of Eskişehir came to Ankara after months of unpaid wages and other employment claims, walking much of the way. They staged a sit-in and hunger strike around Kurtuluş Park. On 21 April, 110 workers were detained when they tried to demonstrate in front of the Ministry of Energy and Natural Resources. On 27 April, police used pepper spray when the workers attempted to march to the ministry. On 28 April, negotiations involving the Ministry of Labour and Social Security, the Interior Ministry and the Ministry of Energy and Natural Resources produced an agreement under which the workers’ wages, compensation and other employment rights would be paid. The workers ended their action and declared victory.
The workers returned to Ankara in June because they said the April agreement had not been fully implemented. On 4 June, the union announced that some payments had been made, but the workers remained in Ankara. On 16 June, a new protocol was signed placing the remaining claims on a payment schedule.
In August, the story widened. Workers from Eti Gümüş joined the Doruk workers in Ankara. Both companies are owned by Yıldızlar SSS Holding. The workers were demanding unpaid wages, severance and notice compensation, and other employment rights. Some Eti Gümüş workers had been waiting for unpaid wages since November 2024, while Doruk workers had claims going back even further. A Labour Ministry inspection had reportedly found that the compulsory unpaid leave imposed on Doruk workers was unlawful and that wages, severance and notice pay were owed.
The August action was more confrontational than the April one. This time, on 10 August, more than 100 workers were detained when they attempted to reach the Ministry of Energy and Natural Resources to arrange a meeting with ministry officials. On 13 August, 82 workers were detained again, together with Bağımsız Maden-İş president Gökay Çakır and organising specialist Başaran Aksu. The workers nevertheless continued their protest. There were repeated detentions during the 18-day action.
On 26 August, Çakır and Aksu were arrested and sent to prison. Çakır had been detained at the union’s headquarters in Soma and Aksu in Ankara, where he was participating in the miners’ protest. Both were accused of “publicly inciting the population to hatred and hostility.” The arrest came on the 17th day of the miners’ protest.
Yet the workers won again. After 18 days of protest, negotiations with Yıldızlar SSS Holding reached an agreement, and on 27 August the workers announced victory. The first instalments of their long-unpaid wages, notice payments and severance reached their bank accounts. According to the union’s lawyer, the agreement covered all of the workers’ claims, with the remaining payments to be made according to the agreed timetable. The decision to end the protest was taken unanimously by the workers.
Thus, in April the Doruk Madencilik workers came to Ankara and won. In June, they came back and won again. In August, workers from two Yıldızlar companies came to Ankara and, after 18 days of protest and repeated police intervention, won again.
The point is not that the workers lost. They did not. They repeatedly succeeded in obtaining what they were owed. The question is why workers had to come to Ankara in the first place—and why, after agreements and interventions involving state institutions, they had to return.
There is another striking feature of the August episode. When the workers declared victory, Gökay Çakır and Başaran Aksu were still in prison. As of this writing, they remain there.
The workers won. Their union leaders remained in prison.
The Institutional Question
The Yıldızlar case brings me back to the question I began with. Akerlof and Romer showed how those who control firms may find it profitable to loot them rather than preserve their long-term viability. Black showed how those who control organisations may subvert the controls designed to constrain them. Veblen, in his analysis of sabotage, showed how vested business interests may restrict productive capacity and efficiency for pecuniary purposes. These concepts help illuminate different parts of the Yıldızlar story.
But they leave open another question: what happens when the institutions that are supposed to constrain private interests are themselves involved?
I introduced the concept of the looting state to provide answers to this question. By a looting state, I mean an institutional configuration characterised by state capture, rent creation and distribution, and the selective use of institutions. The state need not be absent or weak. On the contrary, it may be highly active—privatising public assets, allocating licences, regulating markets, intervening in distressed enterprises, and enforcing obligations. What matters is how these powers are selectively exercised and for whom.
The Yıldızlar case provides an opportunity to see this configuration in operation. Valuable productive assets could be transferred into private hands while obligations were left behind; institutions could intervene without necessarily resolving the underlying problems; and productive alternatives could be restricted in the pursuit of pecuniary interests.
The subsequent struggles of the workers show the other side of the same institutional configuration. They repeatedly had to mobilise to obtain rights that should not have required such mobilisation in the first place. The workers of Doruk Madencilik and Eti Gümüş, whose struggle opens this article, were not asking for charity. They were asking for what was theirs—the wages they earned, the rights they were owed, and the respect that was never given.
The question, therefore, is not simply whether Yıldızlar looted, committed control fraud, or engaged in sabotage. It is whether the institutional arrangements surrounding it made such processes possible—and, at times, facilitated them.
That is the looting state.
References
Akerlof, George A and Paul M Romer (1993): “Looting: The Economic Underworld of Bankruptcy for Profit,” Brookings Papers on Economic Activity, https://www.brookings.edu/wp-content/uploads/1993/06/1993b_bpea_akerlof_romer_hall_mankiw.pdf.
Aydoğan, Metin (2005): Türkiye Üzerine notlar: 1923-2005, İzmir: Umay Yayınları.
Black, William K. (2005): The Best Way to Rob a Bank is to Own One: How Corporate Executives and Politicians Looted the S&L Industry, Austin: University of Texas Press.
Cömert, Hasan and T Sabri Öncü (2023a): “Monetary Policy Debates in the Age of Deglobalisation: The Turkish Experiment—II,” Economic & Political Weekly, Vol 58, No 11, pp 10–13.
Öncü, T Sabri (2022) (2022): “To Loot or Not to Loot? How Public–Private Partnerships Harmed Turkey,” Economic & Political Weekly, Vol 57, No 29, pp 10-12.
Veblen, Thorstein (1921): The Engineers and the Price System, New York: B.W. Huebsch.
Notes
[1] At the time, it was still acceptable to refer to the country as Turkey. Today, even the IMF and the World Bank use the spelling “Türkiye.”
[2] Star Trek
[3] https://www.patreon.com/SabriOncu/posts/documents-for-my-167809675
