Banks Return €137 Billion In LTRO Funds To ECB: Goldman's Take
As expected, moments ago the ECB announced the results of the first LTRO repayment option. According to media reports, a total of €137.2 billion will be repaid as 278 banks participate in the repayment. The consensus expectation was for a repayment figure of €84 billion, so a figure substantially more than both the expected, as well as the whispered goldilocks number of €100 billion. The banks that will free themselves of the LTRO stigma will be disclosed in time - there is no public list, however as a reminder some 523 banks participated in the first LTRO. Since Europe is currently in the risk on phase, don't expect an immediate retaliation against the primarily non-core banks that opt to keep the LTRO funds. The market response so far has been one of risk on, due to the perceived implication that the interbank market is healhtier than expected, coupled with a push up in the EURUSD as the repayment is, as noted previously, a gross deleveraging of the ECB balance sheet coming at a time when every other bank is explicitly devaluaing their currency. Indeed, moments after the announcement the EURUSD ramped up to 1.3460 despite some ugly UK GDP news earlier.
However, the bigger question is how will the market react when a substantial chunk of even inert liquidity is removed, and with it - a backstop to risk, which as Citi warned: if too much LTRO is repaid, the risk goes straight to peripheral bond yields, which have a feedback loop to the currency rate, and thus to overall market risk. Read more in the previously posted Citi note below.
Goldman's immediate post-mortem:
LTRO: Banks retain LTRO funds and return only 13% (€137 bn)
€137 bn putback, €882 bn outstanding
Today (January 25) at 11:00 GMT the ECB announced the amount of LTRO funds the banks returned; banks repaid €137 bn of the LTRO, leaving €882 bn outstanding.
The ECB does not disclose the country breakdown and we will have more colour when NCBs publish their monthly balance sheets. Nevertheless, we expect large part of the putback to come from banks in the core, while we expect limited activity from the peripheral banks. For the latter, the terms of LTRO still make a repayment uneconomical, in our view.
Quantum: as expected
The €137 bn repayment brings total LTRO outstanding to €882 bn. The amount of putback is broadly in line with market expectations, and leaves meaningful excess liquidity in the system.
From here, banks have the option to repay (partly of fully) the LTRO on a weekly basis until expiry. We see a >12m reduction process as the most likely outcome of further banks’ repayments.
Exit structure: an ongoing option
ECB’s LTRO offer was structured as “take it or leave it”. This resulted in two, dramatically high, bank liquidity injections (€0.5 tn each). The structure of the putback option, however, is the opposite, allowing banks weekly repayments – full or partial – over the entire duration of the facility (January 2015).
Signaling and maturity are crucial
Without time pressure, a putback decision is driven by economics. We believe that for peripheral banks, LTRO money still offers an attractive reinvestment proposition.
Moreover, we believe the banks will use 4Q2012 results to outline longer-term path of repayments, in order to signal their ‘resilience’.
Finally, with 2 years remaining, LTRO remains an attractive facility for the majority of banks, that cannot achieve comparable terms in the funding market. But in a year’s time, this is unlikely to be the case. We believe a longer-term exit path will be the most likely outcome.
And, as a reminder from yesterday, Citi's pre-mortem:
ECB will release data on the early LTRO loans repayment tomorrow. The release will help gauge the liquidity needs of the European banking sector and the prospects for a further reduction in Eurozone excess liquidity from here given that the repayment of LTRO loans will likely continue over a period of time. Consensus expectations seem to be centered around EUR100bn. Recent EURUSD resilience appears based on the market's growing concern that LTRO repayments will be larger than expected (thus reducing the ECB balance sheet / tightening more than expected) and driving up the EUR vs the USD (e.g. ECB vs Fed balance sheet). Critically, as Citi notes, the repayment of LTRO loans will free up collateral in the form of peripheral bonds. This seems to be particularly the case ahead of tomorrow given that Spanish and Italian banks were among the biggest borrowers under LTRO’s first tranche. If these banks opt to benefit from the spectacular rally in BTP and Bono markets and liquidate some of their LTRO collateral (and shrink their balance sheets in the process) this could fuel renewed upside pressure on the peripheral bond yields. This could then dampen any EUR upside post LTRO repayment - and as the main carry-driver for US equity performance, could lead to a risk-off switch quite rapidly. So tomorrow's LTRO repayment needs to be Goldilocks - too little and its clear banks have liquidity problems still; too much and the market's reaction could be notably risk-off.
Authored by Valentin Marinov of Citi,
Hopes for sizeable LTRO repayment leave EUR vulnerable
- Markets could be positioning for a larger LTRO repayment amount than the EUR100bn currently anticipated by the market.
- This could mean that EUR would trade in buy the rumour/sell the fact fashion with investors taking profits on euro longs in the aftermath of the release.
- The higher EUR movers across the board the greater would be the negative kneejerk reaction to a smaller than expected LTRO repayment.
EUR remained well supported across the board despite a mixed bag of Eurozone data and renewed indications that Germany maybe unhappy about the latest price action in EURJPY. EUR resilience seems linked to tomorrow’s announcement of the early repayment of the LTRO loans by European banks. With the euro so bid against higher yielding currencies like AUD and NZD and European short-term rates still bid we suspect that markets could be positioning for a larger LTRO repayment amount than the EUR100bn currently anticipated by the market. Corroborating this view seems to be the fact that EURGBP surged after headlined about a large UK lender looking to repay almost all of its LTRO loans in February hit the screens.
Being long EUR may seem like a reasonable bet at present. If the LTRO repayment amount indeed exceeds market expectations and comes in at say EUR150bn, Euro rates and EUR could go higher from here as investors bet on further aggressive withdrawal of liquidity and contraction of the ECB balance sheet (see also Figure 1 and 2). The EUR-funded carry trades put in place on the back of further ECB easing by the ECB could suffer more as a result. At the same time, a smaller than expected LTRO repayment amount (say 50bn) could be in part explained by the fact that this week is only the first of many when the European banks can repay the ECB loans. Markets could thus opt to ignore a smaller than expected number and EUR-downside need not be pronounced.
With EUR having moved to new multi-month highs against sterling and CAD and now trading close to recent highs against USD, JPY, AUD and NZD, it seems that the market may have priced in a more sizeable repayment amount to a certain degree already. In turn, this could mean that EUR could trade in buy the rumour/sell the fact fashion with investors taking profits on their euro longs in the immediate aftermath of the release. This also means that the higher EUR movers across the board the greater would the negative impact from a smaller than expected LTRO repayment amount be from here.
There are other reasons to expect that any EUR upside on the back of larger LTRO repayment amount could be only temporary. As argued in "EUR ahead of LTRO" (below) Friday published yesterday the repayment of LTRO loans will free up collateral in the form of peripheral bonds. This seems to be particularly the case ahead of tomorrow given that Spanish and Italian banks were among the biggest borrowers under LTRO’s first tranche. If these banks opt to benefit from the spectacular rally in BTP and Bono markets and liquidate some of their LTRO collateral (and shrink their balance sheets in the process) this could fuel renewed upside pressure on the peripheral bond yields. This could then dampen any EUR upside post LTRO repayment.
EUR ahead of LTRO Friday
- The LTRO repayment on Friday will affect ECB’s balance sheet and short-term rates. It will free up collateral, can weaken domestic demand for peripheral debt and push peripheral sovereign yields higher. Our results suggest that EUR could be driven by the response of the peripheral bond markets and less so by the response in short-term rate markets.
- Smaller than expected LTRO repayment (less than 100bn) could the short squeeze in Euro FI to a halt and weigh on EUR. The downside need not be pronounced if peripheral spreads keep tightening, however. Larger repayment could support EUR via higher rates but only if the repayment does not trigger renewed increase in the peripheral bond yields.
ECB will release data on the early LTRO loans repayment on 25th of January. A month later, the banks will have another opportunity to repay LTRO loans taken up as part of the second program announced in February 2012. The release will help us gauge the liquidity needs of the European banking sector and the prospects for a further reduction in Eurozone excess liquidity from here given that the repayment of LTRO loans will likely continue over a period of time. Market expectations seem to be centered around EUR100bn. Citi’s economists expect repayments between EUR50bn-100bn.
The overall impact from the LTRO repayment on EUR will go through two channels:
1) The first is the impact of LTRO repayments on the size of the ECB-balance sheet and the short-term Eurozone rates. A sizeable drop in the ECB balance sheet and Eurozone excess liquidity could be seen as supportive for EUR against currencies where the market is expecting central bank balance sheet expansion like the Fed, the BoJ and the BoE. In Figure 1 we plot EURUSD against the ratio of ECB and Fed balance sheets. The historic relationship would suggest that aggressive shrinkage of the ECB’s balance sheet relative to Fed’s could be EURUSD positive. In Figure 2 we plot EURUSD against EUR-USD 2y swap rate differential. If the LTRO repayment fuels further short-covering in European FI this could lead to wider rate spread and thus higher EURUSD.
2) The LTRO loan repayments will also free up collateral in part in the form of peripheral government bonds. In turn, this could weaken the domestic demand for Bonos and BTPs and add to the upside pressure on peripheral bond yields. As shown in Figure 3, the EURUSD rally over the last few months was driven by sharp tightening in peripheral bong yield spreads to Germany. If that process slows down and even goes into reverse as a result of the LTRO repayments, with foreign investors too slow to fill up the gap in demand, this can add to the EUR headwinds. If, however, foreign investors buy peripheral debt at more attractive levels, the impact on EUR need not be pronounced.
1) LTRO repayment of less than 100bn could dampen the latest pick up in the Eurozone short-term rates and add conviction to the view that the ECB balance sheet would shrink only gradually over time. Given the close correlation of late between EUR and short-term rates – this outcome could weigh on EUR. That said, to the extent that the outcome is perceived as maintaining the constructive outlook for the peripheral bond markets, the EUR-negative impact of the outcome need not be pronounced.
2) Larger than expected LTRO repayment could point at further upside for Eurozone short-term yields and support EUR especially against low yielding currencies like USD, JPY and to a degree GBP. Any potential gains could be contained if the repayment amount triggers renewed widening in the peripheral bond yields to Germany.
The overall impact of the LTRO announcement on EUR will also depend on the relative importance of the two drivers – the relative monetary condition in the Eurozone vis-à-vis the rest of the world and the peripheral bond yield spread. In Figure 4 we plot the sensitivity of EURUSD to changes in EUR-USD 2y rate spread and the average Spain and Italy 10y bond yield spread to Germany.
In particular we use t-statistics derived on the basis of regression betas from a multivariate model of EURUSD on the two drivers. The dashed lines in the chart indicate the statistical significance of the sensitivity measures - +1.64 indicating minimum level of significance for positive t-statistics and -1.64 indicating the same for negative t-statistics.
As shown, EURUSD responds positively to wider EUR-USD rate spread (positive t-statistics in Figure 4) and negatively to wider peripheral bond yield spreads (negative t-statistics in Figure 4). The results indicate that the impact of changes in the peripheral bond spreads is now more statistically significant than that of the EUR-USD rate spread.
We think that the decisive driver of the EURUSD response to the LTRO repayment could be the reaction in the peripheral bond markets to the LTRO repayment announcement. Renewed tightening in the peripheral spread in response to smaller repayment amount could help EUR withstand a potential tightening in the EUR-USD 2y rate spread. At the same time, renewed widening of the peripheral bond yield spreads in response to larger LTRO repayment amount could more than offset any positive impact from widening EUR-USD rate spread.