About elephants and ABS

It is 14 months ¹ since the President of the ECB has announced in May 2013 the intention of the central bank to carry out an ABS (Asset Backed Securities) program. The work, Draghi announced at the beginning of June, has recently been intensified. Everybody knows that it takes 9 months to generate a baby and many will know that it takes 22 month to generate an elephant. So the gestation of the ABS program has so far taken 1.5 times that of a baby and, if the program will be implemented by the end of this year, it will have nearly reached that of an elephant.

There are two possible reasons for such a long preparation time. First, the ECB may want to be over-sure to come out with a significant and successful program. Second, the design and implementation of such a program may be particularly difficult. One can only speculate about the first reason, but there are indeed factors making the  design and implementation of an ABS purchase program utterly complicated.

Let me list the four factors that I regard as most important.

First, the stock outstanding of ABS is relatively small, at about 1.5 trillion € (as I presented in a previous post ² ). Of course, as such the figure looks large, however the bulk of it is constituted by RMBS (Residential Mortgage Backed Securities) and, notwithstanding some ambiguity in the answers of Draghi in his last press conference  ³ it is clear that the ECB has no intention to favour the housing sector, which in countries like Germany may already be too lively. If indeed the ECB would concentrate its attention on ABS having SME (Small and Medium Sized Enterprises) loans as collateral, the relevant stock would reduce to less than 150 billion €. In the same direction goes the consideration that a part, probably something like a quarter, of the outstanding stock of ABS does not belong to the “Qualifying ABS” type that the recent ECB – Bank of England paper mentions and that the two central banks want to promote .

Second, issuance of ABS in Europe is, according to the ECB-Bank of England paper, moribund, so not only the stock is low but it is hard to see any prospect of its increase.

Third, also as a consequence of the limited issuance, liquidity in the ABS market is scarce and pricing uncertain.

Fourth, and most important for an institution planning a significant purchase program, it is not encouraging that the problems is one of supply more than one of demand. Indeed in the current conditions of hunt for yield, investors would be ready to purchase much larger amounts of ABS, at current spreads, than issuers and originators are ready to create. Of course, part of the reasons for the dearth of supply is the action of the ECB itself: for fully justified macroeconomic reasons, the ECB provides, including through the forthcoming TLTRO (Targeted Longer Term Refinancing Operation), cheap, medium term funding to banks and it is not a surprise that banks find little incentive to securitise and sell part of their assets and rather prefer using them as collateral for central bank funding.

Given these complications, it is not only difficult to design and implement an ABS purchase program, it is also difficult to guess how it could look like. Still, I think one thing is clear: there is little, if any, relationship between any forthcoming ABS purchase program and quantitative easing (QE). There are two reasons for this. First, as I argued in the post recalled above, the size of any QE deserving this name must be measured in trillions, which is just another dimension than the one relevant for ABS. Of course, ABS could be one (small) part of an overall QE program, but here is where the second reason becomes relevant. Draghi has clearly distinguished an ABS purchase program from a QE like operation that would be taken if inflation would further distance itself from the 2.00 per cent objective ⁴. So, the hurdle to carry out an ABS program is definitely lower then the one to be surpassed for QE.

If an ABS purchase program would indeed not be part of QE, how could it look like? My idea is that it would be similar to the Covered Bond Purchase Program, with a catalytic, market repairing approach, working more like a chisel than as sledge hammer like QE did in  he US, the UK and Japan.

More specifically, I think the program would have the following characteristics:

1. In terms of size, I think something analogous to the sum of the two Covered Bond Purchase Programs is as good a guess as one could make, so less than 100 billion.
2. Even this relatively limited size could not be done, given the market conditions and the selectivity likely to characterise the purchases, over a short period of time. At least 18 or 24 months may be needed. A longish duration could also be useful in giving time to banks to securitize assets, to establish less penalising regulation and let credit demand grow.
3. The purchases could, at the beginning, include a share of legacy assets but should concentrate, over time, on new issuance.
4. The purchase of any single issue should not surpass a certain share of the total of that issue, say 50 percent, to be sure that the ECB is catalysing the market and not substituting it.
5. Fairly precise characteristics should be established for the structural features of the ABS to be bought, following those presented for “Qualifying ABS” in the joint paper with Bank of England and those enshrined in the label of PCS (Prime Collateralised Securities) initiative. ⁵.
6. Given the very uneven distribution of ABS, currently and prospectively, across countries in the €-area, the ECB should resist the temptation to establish national quotas for the assets to be bought. This would be critical to assure effectiveness and flexibility to the program.
7. The effectiveness of the program would be drastically enhanced, indeed without it one can doubt about the viability of the entire endeavour, if also part of the mezzanine and equity tranches would be bought and not only a part of the senior tranche. If this was seen as putting too much risk on the ECB balance sheet, either an external entity (The Commission? The European Investment Trust as part of the EIB Group ? National public institutions like KfW Bankengruppe, Caisse des Dépôts et Consignations, Cassa Depositi e Prestiti) could offer a guarantee or could purchase part of these other tranches.
8. To obviate the difficulty of the uncertain price and liquidity in the ABS market, the ECB could purchase the ABS under the form of Non Recourse Repos, whereby the bank using a certain ABS as collateral for a repo operation can decide, at the maturity of the repo, whether to purchase it back at the preset price or to leave it with the ECB and be freed of the obligation to return the funds obtained. This modality would be useful, however, only if banks could be freed from the risk of the used ABS for capital charges purposes.

In conclusion, while it is not easy to design and implement an effective ABS purchase program, one can see the contour of a program that would not have dramatic, immediate effects but would help in the medium and long run. The contribution would be not only from a macroeconomic point of view, in remedying the impaired ability of banks to intermediate funds, but also from a structural point of view, in helping the €-area financial system to surpass the limitations and risks of an excessive concentration on the banking sector to the detriment of the capital market. So, if quantitative easing is seen by its supporters as a tool to jolt the €-area economy into higher growth and closer to price stability, an ABS purchase program is more like a corroborating diet, gradually reinforcing the vigour of the €-area financial system and economy.

---