The devil lies in the details: The actual policy rate in Türkiye is not 42.5

This article is one of the most challenging ones that I have written for a newspaper or any other media outlet. I will explain the reason at the end, but first, let me begin with the Central Bank of the Republic of Türkiye (CBRT) Monetary Policy Committee’s (MPC) latest decision.

The MPC, raised the one-week repo rate, known as the policy rate, by 2.5 percentage points to 42.5 percent on December 21. This was in line with the expectations of the markets as well as mine.

The repo rate may halt at 45 percent

The last MPC press release, like the previous one, presented clear messages. It seems that there will be another 2.5 percentage points rate hike at the next meeting. Therefore, unless there is a surprise, the repo rate will probably go up to 45 percent and remain there for a while. The press release states that the levels of various loan rates are in line with the inflation target. It implies that deposit rates should rise a little more. I agree with these assessments. The CBRT also announced regulations aiming at keeping the maximum credit card interest rates and credit card commissions intact to protect low-income consumers. Notice that I used the phrase ‘known as the policy rate’ above. I will elaborate on this below; however, first I will analyze what is the likelihood of inflation converging to the Central Bank’s end-2024 inflation forecast of 36 percent.

Energy prices are falling globally

In this context, it is essential to look at the developments in the main determinants of inflation. The steps taken to rationalize monetary policy as well as control the budget deficit, not to reach higher levels, have significantly lowered Türkiye’s risk premium. Within these circumstances, the CBRT has been able to keep monthly exchange rate increases below the inflation rate.

Furthermore, most of the external determinants of inflation have been in the positive direction in recent months. Energy prices have been falling. It was feared that the Israel-Hamas war would push energy prices higher; this did not happen. In late September, Brent crude oil which was close to $95 a barrel is hovering around 79 dollars these days. European natural gas prices fell, although not to the same extent. The interest rate decisions of major central banks are significant in terms of the amount and cost of capital flowing to countries like ours. The Fed and ECB have signalled that their interest rate hike processes have come to an end. The new topic of discussion is when they will start cutting policy rates.

How can the 36 percent target hold?

In this framework, annual inflation is expected to fall to around 36 percent by the end of 2024 in Türkiye. The main issue here is that annual inflation will not go downhill until the end of 2024. On the contrary, it is projected to rise until mid-2024. For this reason, the CBRT emphasizes monthly inflation developments. There is a downward trend in monthly inflation, even though it is opaque in November.

Attention: We are talking about an inflation rate of 36 percent. If it falls to that level after twelve months, we can all rejoice together; nevertheless, 36 percent is a quite high inflation rate by international standards. For example, an analysis of the inflation rates projected for the end of 2023 exhibits that only ten out of 190 countries have inflation rates of 36 percent or higher. In short, even if we reach 36 percent by the end of 2024, we still have a long way to go.

If there is a U-turn

To get that far, we need to keep the same direction. If there would be a U-turn or a detour, we would be yearning for 36 percent, let alone a significant drop from 36 percent. Reducing the likelihood of a U-turn is especially crucial for financing the current account deficit and repaying maturing external debts. The possibility of a U-turn has recently come to the fore again with Moody’s credit rating announcement. It was also seen in the statements made by the Fitch officials to Ekonomim newspaper.

So, what keeps the likelihood of a U-turn on the table? There are two main reasons. The first is the way the previous Central Bank Governor took office. That is the removal of Governor Naci Ağbal, who had been sincerely trying to fight inflation, which had not yet reached 20 percent at the time. The second is the patience of political actors as the inflation will rise to around 70 percent in May-June 2024, twelve months after the start of the rationalization program.

The question of political patience

What will test the patience of the political actors is not only the fact that inflation will rise for a while, but also, for four months now – July-October – industrial production has been falling compared to the month before. The employment rate has been flat since January. A growth rate of 4 percent or even slightly below would be good for reducing the current account deficit. Nonetheless, will there be patience for a ‘low’ growth rate?

If there is no tolerance and there is a U-turn in ‘rationalization policies,’ it will not be possible to control the exchange rate by standard methods. Again, it would be necessary to resort to unconventional measures, like the FX-protected deposit announced in December 2021, which sooner or later destabilize financial stability. All this aside, there is also the issue of what the real policy rate is.

The way to lower inflation

Inflation is determined by aggregate demand and supply. Factors that reduce supply (for example, a higher rate of change of exchange rate than the inflation rate through cost increases) and factors that increase demand (for example, interest rates on deposits, loans, and bonds which are lower than the inflation rate) increase current inflation. In this framework, the way to fight inflation is to influence the exchange rate, deposit, loan, and bond rates in a disinflationary direction. Of course, all these variables are also shaped by expected inflation. Therefore, it is also necessary to reduce expected inflation.

The main objective of most central banks is to ‘ensure price stability’. This is also written in the CBRT Law. In line with this objective, the CBRT sets an inflation target together with the government and uses the policy instruments specified in the law to achieve the target. The main policy tool in this framework is the policy rate. For central banks to fall inflation to the desired level, they need to influence the variables I have just mentioned. How is this expected to happen?

How does the system work?

Here is how it works: Take the banking sector of a country. There is either a liquidity surplus or a liquidity deficit. Of course, one bank may have a liquidity deficit while another bank may have a liquidity surplus. Nevertheless, in the end, when adding them up, one outweighs the other. In Türkiye, there has been a liquidity deficit in the banking system since late 2010. This is also the case in most countries – except for unusual periods. Under these conditions, central banks cover the liquidity deficit of the system by lending to banks.

To borrow from that central bank, banks usually deposit the treasury bonds of that country with the central bank as collateral. At the end of the maturity term, the bond is returned to the bank and the liquidity is provided to the central bank (repo transaction). In Türkiye, the maturity term of this transaction is one week. The crucial point here is that banks pay interest to the CBRT for this transaction. Currently, that interest rate is 42.5 percent.

The expectation is that when this interest rate is increased, banks’ cost of funds will rise, and this will push up their loan and deposit rates. In the same environment, short-term bond yields will go up. If the policy is considered appropriate and meaningful, the risk premium will fall; this fall on the one hand and the rate hike on the other hand will reduce the exchange rate increase (the Turkish lira will probably appreciate in real terms) and expectations will be positively affected.

Can there be interest on a debt that does not exist?

For this to happen, the Central Bank’s policy rate – that is, the interest rate on the funds it transfers to banks – needs to be in line with its goal of fighting inflation. The inflation forecast for end-2024 is 36 percent. The upper end of the forecast range is just over 40 percent. Considering the lagged effects of monetary policy changes on inflation, one could have said that the policy rate announced on December 21 was reasonable, even if it was not high enough relative to expected inflation, given the likely another hike in January.

Yet, we cannot say that. Because since July, the CBRT has not been lending money to banks through one-week repo auctions. The interest rate of something that does not exist cannot be the policy rate. Instead, it transfers funds through Turkish lira swap auctions against foreign currency. With this transaction, banks lend foreign currency to the CBRT and borrow Turkish lira in return. As of December 22, the total swap stock amounted to 1.5 trillion liras. However, on that date, the banking system did not owe any money to the Central Bank due to the repo transaction known as the policy rate.

The real policy rate is not 42.5 percent

The cost of the Turkish lira currency swap transaction to banks is about 0.5 percent below the repo rate prevailing on the auction day.

“So what? One is 42.5 percent and the other 42 percent,” you might say.

The devil lies in the details. The maturity of swap auctions can be up to three months. Whereas in standard repo auctions, the maturity is one week. For example, on the day before the repo rate was raised to 40 percent on November 23 (when the repo rate was 35 percent), and even on the morning of the day the interest rate was raised, there were funds transferred to banks with a maturity of three months at an interest rate of approximately 34.5 percent.

Maybe there are too many technical details; however, the moral of the story is that on December 29th or January 3rd – I chose the days randomly – the real policy rate will be 38 percent, not 42.5 percent. If you think of the funds lent to banks by the Central Bank as the amount of water in a pool, the average temperature of that water (the average cost of funds) will be the weighted average interest rate of the water poured into the pool (funds transferred to banks at various maturities) in the three months preceding that day. And we do not know this weighted average interest rate on funds. Yet, it is the real policy rate.

The real policy rate should be announced

Okay; the CBRT needs to build up its foreign exchange reserves. Therefore, it makes swap transactions. The increase in gross foreign exchange reserves in exchange for swap transactions does not change the net foreign exchange reserves (because it will return the foreign exchange after a while) but let us forget it and say that an increase in gross reserves is also sound. Therefore, if the CBRT finds it inevitable to continue swap transactions against foreign exchange for a while longer, it should at least consider shortening the maturity of these transactions. It should also announce the weighted average swap rate, i.e. the real policy rate. On the other hand, it should not be forgotten that these transactions push banks to borrow foreign currency to obtain cheaper Turkish lira. This is the second moral of the story.

The third moral is that nothing can replace the foreign exchange reserves that will naturally increase because of the de-risking of Turkish lira-denominated financial assets and their becoming sought-after financial assets. This cannot be achieved through interest rate and tax hikes alone; a comprehensive program is needed.

Is the inflated TL fund for the elections?

Finally, let me briefly touch upon a development that makes no sense. On December 21, right after the MPC decision, the CBRT issued another press release announcing that it would start Turkish lira deposit buying auctions. In other words, it will borrow Turkish lira from banks by paying them an interest rate. Here is the deal: On the one hand, you are lending Turkish lira to banks (against foreign currency) and on the other hand, you are buying it back.

I think this is being done for two reasons. First, the Treasury’s Turkish lira deposit account at the CBRT is currently quite high. It will probably withdraw and use a significant portion of it in the coming days for earthquake – and maybe election – expenditures. Therefore, there will be a lot of liquidity in the system. The size of the swap is about three times the size of these deposits. The problem would be solved if the CBRT cut down on swaps, but it cannot; as I mentioned above, it feels obliged to hold Turkish lira swap auctions against foreign currency to boost gross reserves even if it does not increase net reserves. This is the second reason.

The Central Bank’s bumpy road

At the beginning of this article, I said that I do not remember ever having so much difficulty writing for a newspaper (or similar media).

Part of the reason for this is that if “I am overwhelming the reader with technical details”. Nonetheless, the main reason is that the path the CBRT is walking is extremely bumpy and I feel obliged to elaborate on those bumps.