Why was the last TLTRO take-up unexpectedly high?

Last week the ECB published data about the take up of the fourth and last tranche of funding from the second Targeted Long-Term Refinancing Operation (TLTRO 2.0). The data showed an unexpectedly large demand for liquidity from euro area banks. In total, EUR233.47bn was allotted to 474 Eurozone banks.

Although the level of liquidity in the Eurozone remains high, banks enthusiastically took up this last opportunity to borrow money from the ECB for 4 years at negative rates. Prior to the publication of the figures, the Bloomberg consensus (the median estimate of a Bloomberg survey) was forecasting a take-up of about EUR110bn. At EUR233.47bn, the final amount is therefore more than twice what was expected.

When TLTROs were created in 2014, the aim was was to incentivise banks to finance the real economy. In order to do that, the ECB offered banks a new opportunity to fund themselves for 4 years for a volume calculated from their outstanding loans to the euro-area non-financial sector (TLTRO 1.0). In the second programme, TLTRO 2.0, banks were “allowed to borrow an amount equivalent to up to 30% of their outstanding eligible loans on 31^st of January 2016, net of the funds from the previous TLTRO that they may still need to repay”.

The interest rate of the loan is normally determined by the MRO rate at the date of the take-up (0% today). However, if the bank’s net lending between 1 February 2016 and 31 January 2018 is higher than a specific benchmark, the cost of TLTRO can fall below that. Specifically, the rate applied can be as low as the deposit rate at the date of the take-up. The ECB’s deposit rate entered negative territory to -0.1% in June 2014, and three more cuts followed, leading the deposit rate to -0.4% today.

This innovation in TLTRO 2.0 reduces the cost of holding increasing reserves at the ECB because of the negative deposit rate. Indeed, if banks borrow at 0% but need to pay the deposit rate of 0.4% for holding reserves, there is some cost involved. However, if banks borrow at 0.4% and pay 0.4% for their deposits, borrowing becomes costless for them. In addition, expectations about the future path of the deposit rate also play a role. If banks expect the deposit rate to increase over the next four years, they can actually make money out of these loans, regardless of what they do with the money. Take up of this fourth and final round of TLTROs might therefore have been influenced by a significant change in expectations.

Following the ECB’s last press conference, in which Mario Draghi was considered a bit more hawkish than previously expected, and with headline inflation suggesting that the ECB is close to its target (even though core inflation is still far from the target), the market is now pricing in a deposit facility rate hike in the first quarter of 2018. This is an important shift. One year ago, market participants were still expecting a further fall in the deposit rate, while now they expect a rise.

Let’s take a look at the EONIA (Euro Overnight Index Average) forward curve, which gives us a picture of market expectations about possible future hikes in the deposit rate. Figure 1 shows the change in market participants’ expectations after the last ECB press conference. As of 22 March, they are pricing in at least one or two 10 bps hikes in the deposit facility rate over the next 12 months.

Looking at the trend in the 1y1y forward EONIA rate (Figure 2) also shows that markets reacted strongly ahead of and straight after the ECB’s last Press Conference on 9 March. The rate increased by 15 bps in less than a week, confirming that market participants now expect hikes in the deposit facility rate.

On aggregate, German and French banks already met their own net lending benchmark and the 2.5% loan stock target, which means that they should easily get the reduced rate of -0.4% for their take-up of the TLTRO. Italian, Spanish, Portuguese and Greek banks still have to increase their net lending to the private sector, if they want to lower the reimbursing rate and benefit more from the operation. On aggregate, the periphery countries met their negative benchmarks; however, their eligible net lending remains still below the 2.5% of benchmark loan stock. (See the ECB’s latest Economic Bulletin for more information about developments in net lending to private sector.)

To conclude, through the use of TLTROs banks were able to secure funding for four years at a rate which can go as low as today’s deposit facility rate. Given that they expect the deposit rate to go up during these four years, by securing a four-year funding at a reimbursing rate potentially equal to -0.4%, banks are actually forecasting to make money thanks to TLTROs. This might explain why the last TLTRO was more attractive than the previous ones.