Delegate biographies: Learn more about the panelists
- The government response to the 2000/01 crisis fundamentally improved Turkey’s stability, growth prospects and attractiveness to foreign investors
- Non-performing loans and FX risk have been drastically reduced
- Extreme Turkish lira exchange rate volatility seems to have been overcome
- Government bonds remain the most popular fixed-income investment and more needs to be done to encourage the growth of a corporate debt market
- Turkish banks remain well capitalized and mainly funded by deposits
SB, Euromoney Let’s start by looking at the crises of 2000/01 and 2007 and how investors should view the Turkey that has emerged?
The 1990s were a lost decade for Turkey and the main failure of the Turkish economy throughout that decade was governance. The 2000/01 crisis created the impetus for the overhaul of the governance structures of the economy – the independence of the central bank, the overhaul of agricultural support policies, the strengthening of banking regulation, the strengthening of other regulatory authorities and the setting up of new ones. This created the infrastructure that has allowed the country to withstand the current crisis so well.
It also created the foundations for the sustainable growth patterns post-2001. Previously, Turkey’s economy was much more volatile. Yes, we had spurts of high growth, but they always culminated in a crisis in which growth then plunged 5% or 6%. That has changed. That up-and-down cycle of growth came to an end with the radical overhaul of governance structures and has been replaced with high and continued growth. And other areas have benefited. Take foreign direct investment: over the past couple of decades FDI averaged between $500 million and $1 billion annually. In 2007, Turkey had an FDI inflow of $20 billion. One driver of that has been the more benign global economic environment but Turkey was able to attract this investment in competition with other emerging markets because Turkish economic governance structures gave confidence to international investors.
CY, Ak AM
I agree. After the 2000/01 banking crisis the government started a rapid but well-defined financial reform programme. Public banks were recapitalized and made subject to increased supervision. Insolvent banks were transferred to savings deposit insurance funds, and were resolved quickly through mergers, transfers, sales and liquidation. Private banks obtained capital increases and restructured bad loans.
One crucial lesson the banks learnt was the importance of liquidity. Consequently, more than 60% of the banks’ balance sheets are now funded by deposits. As a result I agree that, thanks to the 2000/01 financial crisis, the Turkish banking system has become stronger than many of its foreign peers.
SB, Euromoney Botan, is that how Fitch would characterize the developments, and is that reflected in your analysis?
Yes. Turkey had managed to overcome some of its fiscal deficiencies during the 1990s, such as putting an end to the central bank’s monetization of the government debt, by signing the protocol between the treasury and the central bank. In addition, financing the agricultural sector had been somewhat addressed, and financing by [state grain board] TMO had been stopped. But weakness remained because Turkey had misaligned its regulations (especially the financial sector regulation) with the liberalization policies that were adopted during those years.
At the time of the 2000 crisis the banking sector was betting heavily on interest rate arbitrage by borrowing heavily from foreign markets and lending to the domestic market through buying Treasury bills. By taking a high risk the banking sector was making high profits but in a period of a month when the Turkish lira depreciated sharply all that profit disappeared because of banks’ high open positions in foreign exchange. And outside the banking sector there were problems because of a lack of control over the non-bank financial sector.
So I would see the 2000/01 crisis as a mismatch of the regulation and liberalization policies. There was a mismatch between the pace of those two, and right after the crisis this changed. [Bank regulator] BRSA started to function as an autonomous institution and immediately issued regulations on the recapitalization of the banking sector and redefined the risk structure of the banking sector. The strength of the banking sector today was achieved during the crisis times of 2000/01 and so, yes, the crisis is a key reason why today Turkey is standing up better than many other countries.
I will approach the situation from a different view. After the 2000/01 crisis we did something that made the banking system and the other financial systems more conservatively structured: we let some banks fail. And yes, we also imposed a more conservative regulatory system for the banking system and for the financial system. We also changed the economic governance of the state, the governance structure of the Treasury, and the system. And all these were very significant.
However, what is more important is what has happened since 2003: political stability in Turkey, and the funds surplus outside Turkey. These two things were more crucial in putting the banks in Turkey into a better position than others in the latest crises.
SB, Euromoney Ufuk, you were in risk management at the Treasury, how do you think things have changed between that previous crisis and this one, and where do you see the most improvement?
UH, Turkish Treasury
I agree with what has been said already but, if anything, what they have underplayed is the sectoral and structural transformation performed. In 2001 the non-performing loan ratio in the banking system was 21.3%; in 2009 it is around 3.9%. In 2001, the entire balance sheet of the banking system had exchange rate risk of $5.4 billion, now it is around $200 million. The average capital ratio during that first crisis was around 5%, it is now 17%. So, structure-wise, the Turkish financial sector is much more developed. We did not even have any risk management activities during 2000/01. Starting from 2002 with Law 4749 we established a risk management unit within the Treasury and now it is one of the main essentials of the private financial sector as well.
But, on the other hand, there is also one big difference between this crisis and the last one. Then, Turkey faced financial crisis on her own but now everyone, every country, is facing this crisis. So the market is saying to us: ‘Look, you have to deal with your own internal dynamics – but with the external dynamics as well.’ So the advantages and experiences of 2000 and 2001 are substantially important – but not enough. As governors and economic agents we have to prepare ourselves for unexpected risks that we have never faced before.
SB, Euromoney And the government itself has reduced its risks – for example with foreign-currency borrowing?
UH, Turkish Treasury
From a risk management point of view we need to take care of market risk and liquidity risk. To deal with market risk we have to borrow as much as possible in Turkish lira, and we have to borrow as much in fixed-rate terms as possible. Otherwise any interest rate fluctuation or FX rate fluctuation might hit us in coming years. Going five or six years back, FX-denominated debt was about 32% of our total domestic debt in 2002. That is now down to about 6%.
That was perceived as a risk by the rating agencies, and we think that risk has been reduced over the years by issuing more in Turkish lira.
SB, Euromoney To what extent are your ratings driven by these broader variables, such as the systemic strength of the banking system and your perception of the strength of the regulatory and compliance systems? And how does Turkey compare with other emerging markets that you have come across?
That is a good question. To some extent ratings take a granular approach. But at the end of the day everything boils down to the default probability, so whatever could contribute most to a default of an entity would have the highest weight in a ratings model. So, if systemic weakness were judged serious, meaning that a high probability was assigned to the banking sector’s fragility to lead to a need to access public funds and in turn causing the sovereign difficulty in servicing its own liabilities, then the strength of the banking system would be significantly weighted in that sovereign’s rating.
SB, Euromoney And are there political risks to the achievements that have been made? Is there political stability?
Well I think there is also a question about fiscal stability in the shorter term. But to answer your question: we asked previously whether the gains that we have seen in this country so far are reversible or not. There the IMF factor, if ever there is an IMF deal on the table, will be a very strong element to convince the international community that these gains have been consolidated, and whether we will see further gains down the road.
Very interestingly we have been talking about this issue for almost two hours and the reference to the EU only came around after half the discussion, which I think is very telling. It shows how much leverage the EU has lost here whereas up until a few years ago we were talking about the twin anchors for the economy – the IMF on the one hand and the EU on the other. Now we do not have the IMF on the table and the EU is very much weakened – so much so that in the past few days the government launched a very important package regarding its industrial policy in terms of the new incentive programme that will be made available. That programme is incompatible with EU rules because it gives sectoral incentives; so even that shows how much of a delinking has occurred in economic policies between Turkey and the EU.
However, I would like to go beyond that, because in the coming months there will be another element that will change the political risk of Turkey, and that is the new round of elections. In all likelihood as we move forward towards election time the new government in Turkey might be a coalition government – and we have bad memories of a coalition government in terms of rent distribution and economic populism. Now, on the positive side I have to say that those bad memories relate to a different time, because at that time of bad economic policies they were put in place when the institutions were weak, the governance institutions were weak. Now it is a totally different set-up, so even if we have a coalition government, that may not necessarily mean that the economic policies will be the types of policies that we have seen in past coalition governments. That would be my first point.
The second point is that when you look at the political discourse it has fundamentally changed. The Turkish electorate has matured, so there are very few buyers (I would tend to think) of the type of open-ended economic populism that we used to see. So not just the opposition parties, when they criticize the government about the type of electoral promises that are being made, but also the electorate will ask: ‘Where will you find the money? How will you do that?’ and so on and so forth. That also needs to be put into its own context, that electoral politics in Turkey have matured. Thus we may not be so afraid of a coalition government as we used to be. Nonetheless I think the prospect of having a coalition government will affect political risk, and therefore yields and therefore the development of the corporate bond market.
SB, Euromoney To what extent? Because the danger is of an abrupt shift in rates or FX in the event of an unwanted political development?
UH, Turkish Treasury
Going back a few years to 2000 and 2001 we were running a completely different exchange rate system at that time, so I am not expecting anything like the kind of jump we saw before where, say, the rate is 1.5 [to the dollar] today but tomorrow morning is 3.0.
Not at all.
UH, Turkish Treasury
Not at all, no. We are not expecting anything like we had before. If demand for foreign exchange increases, it does not mean that we will be facing doubled or tripled rates the next morning, because the system will not allow that. During the day, it will move up or down and the central bank will manage the excessive volatility. That is how they do it for the time being; they announce: ‘If we observe any excessive volatility due to speculation or external shocks, we may intervene, which would be consistent with a floating exchange rate regime.’ So I do not think that procedure would change even if any unexpected and unwanted political development occurs.
Why an IMF deal?
SB, Euromoney Given that right now the strength or weakness of banking systems is a good proxy for overall strength, if Turkey’s banks are so healthy, and if there is political stability, why the need for an IMF deal?
Any IMF deal today would be completely different to the previous deals. All of those were necessary because of fiscal problems or a deep balance of payments crisis. Now the reason for having – or not having – an IMF deal is totally different. Turkey has ample reserves and even the export/import ratio is improving, for example. Today if a deal were done it would be to counter the negative impact that the global crisis has had on Turkey. All the past deals were essentially as a result of a domestic crisis. This is a fundamental difference.
Also, the other crises were about survival. Now the stakes are not that high, so the government has the luxury of deciding whether they wish to have an IMF deal or not, which was not something that Turkey could afford in the past.
I agree. There is no payments crisis. There is a global crisis of which Turkey is taking its own, relatively speaking, small share. Last year when we were looking at the financing requirements of the Turkish economy, the current account deficit that we had in mind was much larger than that which will be realized this year.
The current account deficit is expected to shrink to somewhere around $10 billion in 2009. Banks have refinanced more than 70% of their due loans and in the corporate sector we do not see any refinancing problems, they have been able to refinance. That is again around the 69% to 70% level, and the government is completely performing its own debt servicing, so clearly we do not see a refinancing problem in 2009 and do not think that an IMF loan is imminent in that perspective. Recently we also made an announcement that a signed deal or an unsigned deal would not change our ratings on Turkey.
UH, Turkish Treasury
We do not see any risk in terms of financing of domestic or external debt. Our spreads are moving in line with where we would expect relative to our peers. We are seeing the fluctuation as quite normal considering the global events that we face today and afterwards.
The IMF issue has never been put into our financing programme at all. We normally propose our annual financing programme by the middle-to-last quarter of every year to top management, and we did this in November 2008, and at that time there was not much discussion in terms of the IMF.
The market knows that the current financing programme does not contain IMF provision and has set its perception accordingly. We do not observe any hesitation and/or doubt from the market that refinancing would be a problem without the IMF.
CY, Ak AM
Although we all know very well that the borrowing requirement of the Turkish Treasury is increasing we believe that an increase in domestic debt issuance can be absorbed mostly by Turkish banks. Although I agree that it is not a must, it would perhaps be good to have another IMF deal with a larger package to have a financing buffer.
SB, Euromoney And isn’t it true that the discipline imposed by an IMF deal would reassure investors and cement the existing reforms?
That is an extremely interesting question, because many people do indeed argue that the root of the transformation of the Turkish government in 2001 was the model that was put into place by the previous government with IMF support. Going forward the main risk for Turkey if we talk about the debt market will be fiscal sustainability and fiscal stability. What we will face will be a period of lower growth, which we have not been used to in this decade, as all the fiscal successes of the Turkish economy were achieved at a time of high growth, at a time of benign global conditions, at a time of high liquidity and high growth.
The period ahead of us will be a period for the world of lower growth, and most probably for Turkey as well, so Turkey will need to shift its growth model that has been so successful in the past. This is for me at least a big question mark, because what Turkey failed to do in this period even with the IMF was to address the fundamental pillars of its growth model, which are structural reforms. When you look at the track record of the government it has been extremely successful on the fiscal side but less so on the structural reforms that are needed to have a more sustainable growth model for Turkey.
We are talking here about the informal economy, about the labour market, about the energy market, about infrastructure, investments in transport and so on and so forth. From that perspective the question is a very legitimate one, whether having an IMF deal that would focus on these structural reforms would help Turkey, and I think that the answer is yes.
SB, Euromoney What actual evidence do we have for the investor view of Turkey?
UH, Turkish Treasury
So far we have not faced a demand problem and in fact we are seeing decreasing interest rates. If you look at the year-end inflation forecast the central bank is expecting a decrease in the inflation rate and a possible reaction to that if it continued would be lower interest rates. So then I would have lower spreads and lower absolute borrowing costs.
We have seen the foreign share of outstandings fall slightly – it used to be 10% to 13% a year and a half ago and it is about 9% now – but that is not significant. Most of the money that left the country was hedge funds, not real money.
SB, Euromoney And are domestic investors still happy investing so heavily in domestic government paper?
CY, Ak AM
As Ak Asset Management we are managing the mutual funds of Akbank, the pension funds of Aviva and what we call segregated accounts for ultra-high-net-worth individuals and institutions. Due to the favourable demographics of the population, at an average age of 29, Turkish people are at the stage of accumulating wealth in their private pension accounts. I estimate that 90% of these assets will be invested in Treasury bonds and money market instruments. Turkish investors like the safety, they search for high yield against low risk, and they like simplicity. So I believe that the domestic investor base will be able to absorb government issuance in the mid to long term, assuming that the government will be pursuing a tight fiscal policy.
SB, Euromoney It does seem though that the government is crowding out other borrowers. Could more be done to encourage a corporate bond market?
Especially after the 1994 crises until 2006, there was a very small amount of corporate bond issuance in Turkey. Before that, there were some corporate bond issues, mainly through on-balance-sheet securitization of the receivables. After 2006, we have experienced some corporate bond issues (around TL850 million) from financial institutions, mostly by private placement to the banks. There are a number of reasons for the slow development in this area. The most important one is the macroeconomic situation, the public was in the need of borrowing. Therefore the crowding-out effect was a major reason. Another important reason is the tax and other costs on the corporate bonds. There was a different tax approach to the government bond investors and corporate bond investors. Corporate bond investors had to pay more tax than investors in government bonds. This different approach was lifted after 2006. Today, the only difference is the banking and insurance transactions tax that the financial institutions holding the corporate bonds have to pay.
Other reasons for the slow development in the corporate bond market are the length of the execution process, which pushes issuers to the bank market, and the lack of transparency in terms of financial data, especially for unlisted companies. And there is an issue with the level of security of the corporate bond holders.
Yes, this is a key problem. In Turkey we do not have a definition of the different receivables in the case of bankruptcy; normally a bond is ranked above all other receivables in case of bankruptcy, and in Turkey it is defined among those at the very end of the ranking scale. This is a major obstacle for increasing demand for corporate bonds.
SB, Euromoney In other words there is no difference between senior and subordinated? A corporate bond is just subordinated?
It is just above equities and that makes the rating of a bond lower than the rating of the entity itself, so it is a major obstacle for corporate bonds.
However, there is good progress in the corporate bond market. As I said before, the tax differences between government and corporate bonds were removed. In January 2009 the Capital Markets Board of Turkey made some changes in the regulation of the corporate bond sector. The key change is that the debt instruments are now regulated under the same communiqué with common, principle-based rules. The umbrella communiqué defined the corporate debt instruments in more flexible structures. We also redefined private placements in greater accordance with EU regulations, which again has made the issuing process easier. And we have increased flexibility in maturities. Previously, corporate bonds had to be between two and seven years. The maximum term of corporate bonds is now unlimited, and the minimum is one year, which is an important issue for the bonds. And it is now possible for issuers to issue any kind of debt securities without enacting a new legislative provision. For example, a company can bring us a structure that has the characteristics of a debt instrument. The board can say: ‘This is the debt instrument and can be registered with the CMB.’
In addition, in order to lessen issuance costs, all securities, whether offered to the public or private placement, have been dematerialized within the central registration agency. We are also working on changing the CMB registration fee for the corporate bonds.
Equity-linked borrowing is gaining importance after the crises. There are still some problems with issuing convertibles, therefore we structured exchangeable bonds (with the shares of another listed company) as a new type of instrument for the companies that seek to reduce the cost of borrowing.
Are these measures enough to improve the corporate bond market? I think that further improvements can be made. Reducing other fees, allowing corporate bonds to be eligible for repurchase agreements and collateral in the loan market, allowing the mutual funds to invest in corporate bonds are some of them.
SB, Euromoney What about foreign-currency-denominated bonds?
Companies are not allowed to issue foreign-currency-denominated bonds because it is forbidden by the central bank. Clearly excessive foreign borrowing is undesirable, but foreign-currency-denominated bonds can offer companies additional hedging and financing benefits.
Ratings for all
SB, Euromoney A lot will depend on whether or not a sufficient number of companies can be rated. How is that going? Can you rate the smaller companies?
Yes, of course we can. Our rating covers more than 20 non-financial-sector companies in addition to our other ratings, such as almost all banks, some municipalities, non-bank financial-sector companies and asset managers in Turkey. We are rating mostly the large corporates currently but there are some relatively small corporates in our portfolio. On an international scale, Turkish corporates are relatively small.
SB, Euromoney Are those requested ratings?
They are all requested ratings. We do not do any unrequested ratings in Turkey. Actually some of them had plans to issue abroad and some have already issued abroad, and some are waiting for the market to develop domestically so they have started a rating relationship with us. For some companies, they have bank relations in which banks ask for ratings from them and they use it for their transparency issues, and it works well for those corporates.
SB, Euromoney Is the Treasury in favour of a deepened developed bond market, or are there two sides to this story? The more successful large corporate bond issues are, the less money is available for the government.
UH, Turkish Treasury
We are very much in favour of having another market in addition to the sovereign bond market, in widening and enlarging the market environment. We do not see the corporate bond market as a competitor of the government bond market.
There will always be a conflict there, but nonetheless I think that going forward there is very much a real benefit for Turkey to deepen its financial markets, first of all from the macro perspective when we talk about the non-bank financial institutions, whether they are insurance companies or other companies. The growth of that segment will enable Turkey to perhaps overcome its low savings rate dilemma. As we know one of the current difficulties that Turkey has had in the very recent past, not so much with the ongoing economic crisis as the current account deficit, is a structural issue. Essentially it boils down to a low savings rate as opposed to a high investment rate.
When you think about how the savings rate can be increased, one of the policy suggestions is to deepen the financial market by way of having non-bank financial institutions, and when you have those institutions and they are active in the marketplace, when they sell policies, that will in time lead to a higher savings rate, so there is a macro benefit in it on the one hand.
Now, on the other hand even today if we had a more liquid – or even if we had a corporate bond market – that would have allowed some of the companies in Turkey to have access to finance. The problem is, as you rightly pointed out, that banks in Turkey have been very active and they have been able to give that to whatever company they wanted to. Nonetheless there is a limit as to what they can do.
What the banks have done in the recent past, and I am not a banker so I am more at ease in talking about that, is cherry picking at times of crisis. They have been able to finance some of those companies but they have not looked further down, a line below the top or the blue chips and so on, in times of crisis. In normal times that must be easier, so when you have a corporate bond market that would direct some of those blue chips to the corporate bond market, so that bank debt could be leveraged for the need of the SMEs and perhaps the more risky companies in Turkey, and that is the way you devolve and leverage your economy. I think on both sides there is a benefit for the government, but fundamentally yes, there is also a dilemma between the two.
Turkish residents have been free to buy any kind of foreign-issued bonds since 1989 but they keep on buying Turkish Treasury bonds. Although they can buy any kind of corporate or any other sovereign, they do not; they prefer investing in Turkish Treasury. It is regarded as risk-free in Turkey and it has a good yield.
CY, Ak AM
From the micro perspective, I think the most important issue is the investment philosophy of Turkish investors that characterized itself as short-termism and extreme loss-aversion after experiencing several crises in the past decade. If you look at the composition of Turkish household investments, you can see that the share of deposits in total household investments is around 75%. Due to the recent crisis it will be much more difficult to change the investment philosophy of Turkish investors in the near term. Hence, we can say that the basic products will stay on top of the popularity list for now.
In the long term, when the equation changes in favour of riskier products, Turkish investors’ philosophy will change – hopefully. We believe that when the interest rates on long-term government bonds come down to around 10%, Turkish people will prefer to invest in corporate bonds with higher spreads.
Well-diversified corporate bond mutual funds will be channelling the household savings to the real sector to provide funding for Turkey’s economic growth. As Ak Asset Management we are very much aware of the fact that we have to provide our clients with high-quality investment advice to facilitate this change in investment mentality.
This conservative behaviour was of course a product of volatile times as well, when we had very high real interest rates on demand deposits. Now those have come down significantly, and therefore the value for an investor of holding its assets in demand deposits at 12% nominal interest rate has diminished, so that sort of creates the ground for people like Mr Safak to come and convince us about the usefulness of the new products that will be launched in this market.
UH, Turkish Treasury
This is all true but I reiterate that the main issue is to change investment behaviour. We have to explore corporate and other alternative bond markets and explain mechanisms in those to investors much better.
You are right in addressing the issue of changing investing behaviours, but that is true for individual investors. We should also consider institutional investors, and especially the insurance sector, which is very small in Turkey.
I think one area of focus should be developing the insurance market, and insurance companies will increase investment in other products and longer-term products. They will change the whole scenario, and I think that would be a way of increasing the demand for corporate bond markets. They can look for long-term higher yields for their own clients, so that would be a way to increase demand, otherwise just focusing on individual investors will not lead us anywhere.
CY, Ak AM
Although in general I agree with you, the definition of individual and institutional investor may be confusing. For instance, although I am from the institutional investor side, we are managing mutual funds and pension funds on behalf of our individual clients, we also see that institutional clients are interested in new investment ideas. I think in Turkey we need to increase the attractiveness of long-term products in a synchronized way for both institutions and individuals.
SB, Euromoney No one has mentioned yield: if corporate bonds yielded enough, then people would buy them in preference to governments. Presumably the spread is not high enough.
UH, Turkish Treasury
Yes, that is one of the reasons that we are having higher rollovers than we planned – maybe not the first and major reason but one of the reasons, because investors do not have any other alternative giving the same yield or the same security that they could get.
SB, Euromoney Finally then, since the banking system is so crucial, what are the prospects for the sector in the coming year?
We think the prospects are good. Capital adequacy ratios in the high 17s are backed up, more importantly, by the quality of the capital. Turkey’s banks have low levels of upper tier 2 and have significant free capital. Asset quality is high, and coverage ratios are high; and risk management systems are BRSA-approved. Furthermore, the banks are mostly funded by deposits.
SB, Euromoney Prospects for earnings?
These are also currently good, because we are in a falling interest rate environment where net interest margins are rising. Regarding profitability, the structure of balance sheets is different in Turkey with more than 20% of assets in government bonds. These are high yield and give a liquidity cushion to the banks. It also reduces their vulnerability to asset-quality problems.