The Kapali Carsi

Turkish LIBOR (TRLIBOR) Is Safe!

Here’s guest blogger Ali G. (not the rapper, yeah yeah, I have to make this joke every time) again:

Libor – London Interbank Offer Rate- is a set of rates for different currencies and maturities that is supposed to be the benchmark for some 800,000,000,000,000 USD (that is USD 800 trillion) worth of transactions according to the BBC. A huge scandal erupted a couple of months ago surrounding these rates and it does not seem to wane. First, let’s look at how these rates are determined and how the scandal developed. And then let’s see if we should be afraid of a similar scandal in Turkey regarding our own “benchmark rates”, TRLIBOR.

Libor is controlled by the British Banker’s Association (BBA) and is determined as follows: Every day, a group of leading banks submit their rates to the BBA.  Note that these rates do not reflect any real transactions. They are only estimates of what each bank would pay other banks to borrow in a given currency and maturity. After the rates are submitted, the top and bottom quartiles are dropped and the rest of the rates are averaged. That’s it. So you see in theory it is very easy to rig if the banks in question cooperate somehow.

The scandal started with Barclays. Apparently apart from trying to profit from some of their positions (I will spare you the details) Barclay’s also submitted lower rates in order to look financially sound during the 2008 crisis

Even though both the CEO and the chairman of Barclays resigned following the scandal, the crisis continued to spread from a lot of different fronts. The one that got my attention was a piece I saw in the FT last week regarding Korea’s own benchmark rate scandal.  This got me thinking of our own benchmark rate TRLIBOR.

As far as I know, TRLIBOR is determined through more or less on a similar process, and the Banks Association of Turkey(BAT) maintains it. So are we also in danger? Well, this is actually not the right question. The right question is does it really matter? I know from word of mouth that TRLIBOR is rarely used among banks, even for derivatives contracts let alone credit agreements (floating rate TL credit is very rare). Luckily the BAT also publishes the volume statistics as participating banks are required to submit transaction statistics as well. Let’s look at how deep the TRLIBOR market was in 2012 for each maturity from overnight (O/N) to 1 year (1Y), for number of transactions:

and volume:

Things to note here: First of all, these figures are a joke when compared to the size of the banking system. Moreover, notice that most maturities don’t have any transactions on most days and on some days there are no transactions at all! And finally most active maturities are the shortest maturities, overnight being the most active by far. The short maturity is also not a good sign, as I think the most active are 3 month Libor rates for USD and 6 months for Euribor.

So, if I am not mistaken (please correct me if I am, as I have doubts about the transaction data) it is safe to say that we should not be worried about a rate-rigging scandal in Turkey. This is not because we have a better system to set rates but because nobody seems to care about TRL benchmark rates. I hope one day the market grows to a point where a potential scandal has some chance of making it to the front page.

Ali Gökhan is the acting economist at a Turkish conglomerate. The views expressed here are his personal views only and do not represent the views of his company.

OK, here is your friendly neighborhood economist with his usual post-post comments: First, for some fun, I should note that 800,000,000,000,000 USD reminded me of Dr. Evil🙂 Coming to more serious matters, I don’t have the figures for the amount of variable-rate lending (or share in total), but I can tell you that, for the planned renovation on my moonlighting business, I am being offered variable-rate loans in TRLIBOR. So they live!!!!:) On a related note, the volume and transactions data are not really relevant: What really matters is the size of contracts tied to TRLIBOR, on which I will get back you once I find some data. In the meantime, let me share with you the TRLIBOR rates. First, for the overnight:

Then, for three-month…

And finally for one-year contracts:

As you can see, the spread on the overnight has been very volatile of late. Other than that, except a few spikes, which are probably data glitches, there doesn’t seem to be a huge problem with TRLIBOR. But this is just a casual observation. To determine the optimal loan for our business, I am also trying to do some empirics on the determinants of LIBOR. Once I have something concrete (periods of large deviation from my model, for example), I will have an addendum, where I will report on the size of the TRLIBOR-dependent market as well…

3 Responses to “Turkish LIBOR (TRLIBOR) Is Safe!”

multipleequilibriaAugust 22nd, 2012 at 6:53 am

A few additions I wanted to make on top of Emre's comments:
– I am not aware of any official fixed vs floating credit data in Turkey. If it exists somewhere please do let me know I would be very much interested.
– Therefore as far as I know there is no way to calculate the real volume of floating TRLIBOR indexed credit. However we can do a back of the envelope calculation as follows: From word of mouth figures I would guesstimate that the outstanding notional on floating TL credits in the system would not exceed 750 mio TRL. Now, the banking system as a whole has around 550 bio TRL of total TL credit. I believe this number reduces to around 400 bio TRL when we exclude participation banks and credit cards. That takes the ratio of floating TL credits to the whole TL credits to around 0.2%. I have no idea about the amount on derivatives, but again I doubt it is very large.



ErenASeptember 19th, 2012 at 8:26 am

Couple of things:
– Trlibor volume data is not accurate. Banks “should” report their executions but there is no sanction for understating. Thats why only some of the banks report all deals.
– Trlibor is not active since “unsecured” lending intra banks is still a taboo in Turkey. Total credit lines is no more than 3-4 bio TRY; how could be the market outstanding higher than these lines? There is the covered/secured repo market of 10+ bio and another 10 bio of CBT money. For most of conservative bankers, thats enough. Trlibor means “place where you shouldnt let go money cheap to competitors” to many. So many players just look each other’s rate (during market hours) and alter their quotations to avoid actual transactions.
– The biggest reason for all of is the high liquidity premium in Turkey. Deposit rates are CBT policy rate plus 400 bp! No one wants downsize by letting deposit holders leave. No one wants to lend away that pricey money to competitiors with 200+ bp loss. So, no interbank lending in significant size.

ErenASeptember 19th, 2012 at 8:26 am

– Finally, Libor-linked loans are rare, true. But loan-only appreach would be too simple. The importance of (current) TRLibor is via “swap” market. Interest-rate swaps, which are used to hedge fixed-rate consumer & consumer loans, have a float leg paying 3m TRLibor. This brings out the fact: locals need continuous lower 3m rates to bring down IRS rates gradually, but doing so without throwing out funds to other TRLibor banks ultra-cheap. Whereas, foreign banks, most of whom bring “hot” money to the country via swaps, need “higher” swap rates to taste the juice (rate differentials). So, they may force their “friends” (like a “sub”way) to keep 3m rate high. So, not other tenors but 3m rate is significant for all of the market. You may argue that this is still a short tenor and cannot reflect anything to long swaps. But it’s like the sea carving out the rock: if you keep that rate significantly (lets say 50 bp) lower than what it should be for a rather long time; swap trader will receive/pay 50bp lower float legs continuously. That will be a fact and will change the swap pricing mechanisms.