Primary bond auctions do not typically fail but that does not mean investors should ignore the refinancing needs of sovereign issuers. On the contrary, the high level of volatility in sovereign bond markets and the absence of a Plan B should be cause for concern and encourage investors to remain cautious on European sovereign risk.
A couple of months ago, a sovereign official of a large European country at the heart of the eurozone sovereign debt crisis decided to embark on a “roadshow” around Europe with the aim of convincing the investment community of the long-term sustainability of his country. He set out to explain how the high yields of his country’s government bonds were in fact an opportunity to enhance profits rather than a risk for future losses.
This direct engagement with financial market participants, together with early reform measures, did indeed help to re-establish the sovereign’s lost credibility among international investors.
However, the interesting part of the meeting was not the speech itself but the question and answer session that followed. During this session, a fund manager asked the country official the following question: “Do you have a Plan B should a government bond auction fail?” By way of response, the official turned the question over to his colleague sitting to his left: “Can an auction really fail?” The response: “Of course not. These auctions are designed to never fail.”
This (non-)answer is technically correct: primary auctions are in fact designed not to fail. Unfortunately, this does not mean that investors should ignore the refinancing needs of sovereign issuers. On the contrary, as I will proceed to explain, even in the current market with tighter spreads and ample liquidity, the high level of volatility in sovereign bond markets and the absence of a credible Plan B should encourage investors to remain cautious on European sovereign risk.
To understand why investors should be cautious even if auctions do not typically fail, it is worth spending some time exploring the mechanics of primary auctions. Let me start by saying that rules for primary auctions are not the same for all issuers (Germany being a notable exception among large issuers in Europe), but generally speaking an auction is a market event that occurs over a short timeframe during which risk is exchanged between the issuer (the seller) and a restricted number of market participants (the buyers).
The number of market participants (primary dealers) who are allowed to buy bonds in primary auctions are usually limited. Primary dealers have certain obligations, such as guaranteeing a minimum level of demand at auctions. In exchange for their participation they are rewarded with incentives. For example, primary dealers may be incentivized via “greenshoe options” in select markets. These options give primary dealers the right to buy a pre-defined overallotment of bonds at the same price after a predefined number of day(s) following the auction. The overallotment is a function of the quantity of bonds bought in previous auctions. Therefore, the more primary dealers are willing to buy at auctions, and if they do so consistently, the more potential they have for additional profit from exercising “greenshoe options”.
But there is a catch. Auctions are events where a large amount of market risk in the form of regular price movement is transferred from the issuer to the primary dealer over a short period of time (several billions of euros within a few minutes is not uncommon). In such an environment, the dealers must be able and willing to do one of the following things: (a) take this risk on their own book, or (b) collect buy orders among their own clients to pass through the bonds bought at the auction, or (c) dilute the impact of the auction by selling bonds to their clients in the days/hours before and after the auction. Usually dealers tend to use option (b) and (c) given their role is to facilitate the transfer of risk as opposed to taking on large outright directional risk. In a stable market with low price volatility and large trade flows it can be relatively simple to execute options (b) and (c), but in a market characterized by very small trade flows and high price volatility the only available option may be (a), where dealers have to take the risk on their own books. Even in times of stress, due to the strict relationship between sovereign issuers and dealers, steps are taken ahead of each auction to make sure that the auction itself goes as smoothly as possible. The issuer may start deviating from a predictable issuance calendar and behave more opportunistically, selecting bonds to issue while reducing overall sizes to facilitate the dealer’s job. At the same time dealers must be willing to accept the risk of having to absorb new issuance for a while. This is why auctions do not fail – demand from dealers will still be there even when there is no client demand or flows are almost non-existent. However, a protracted crisis period could result in risk accumulating on the dealers’ books until a breaking point is reached which could potentially result in the issuer being asked to stop issuance altogether.
We have seen this scenario play out three times already since 2010 (see Figure 1). When investors worry about risk of principal losses in government bonds and spreads widen, fewer investors are willing to trade those bonds, and prices become more volatile. As shown in Figure 1, the increase in volatility can reach a breaking point when dealers are no longer willing to absorb risk and the issuer loses market access. When observed from the outside, the shift from having market access to losing it is abrupt and does not necessarily involve failed auctions as a requisite.
Figure 1 also shows how alarming the market conditions were at the end of 2011 when volatility in Spanish and Italian bonds were trading at levels that resulted in a loss of market access for the smaller peripheral countries. It was mainly thanks to the introduction of the European Central Bank’s three-year Long-Term Refinancing Operation (LTRO) that a market collapse was avoided.
The LTRO offered multiple levels of market relief:
- It cut the risk of a bank failure in Europe;
- It stemmed large price drops in risky assets by reducing the pace of deleveraging. Deleveraging will still happen but it does not need to be done via a firesale since banks now have access to long-term funding;
- It allowed willing banks to engage in the “carry trade”, i.e. buying government bonds from peripheral European countries with historically cheap ECB loans;
- It stimulated “animal spirits” incentivizing investors to participate and buy spread assets (government and corporate bonds).
The result has been a massive tightening of spreads across most sovereign issuers. Yields on two-year Italian government bonds went from 7% in November to 1.4% as of mid-March. Meanwhile the Itraxx, a proxy for corporate spreads in Europe, went from 205 basis points in November 2011 to 125 basis points as of mid-March 2012. However, while peripheral government bonds and risky assets from core eurozone countries both benefited from contracting spreads, information on trading flows have indicated differences in investors’ appetites for such securities with risky assets from core countries benefiting from wider support while government bonds in peripheral eurozone countries benefitting almost exclusively from demand driven by domestic investors. Banks, and indeed other types of investors outside of Italy and Spain tended not to
participate in the rally of Italian and Spanish government bonds (see Figure 2).
Uncertainty remains a headwind
Let me now go back to the question posed to the sovereign official: ”Do you have a Plan B should a government bond auction fail?” By not answering, the official implicitly admitted that there is no Plan B in case of failure. Recent comments by European policy makers referring to the fact that the worst may be behind us are along similar lines. It is a huge mistake to be complacent at this stage.
It is true that the LTRO has helped avoid a collapse, but it is like a medicine that eases the symptoms of an ailment rather than fights it. Similarly, a monetary union among heterogeneous countries with no fiscal union is a terminal illness. Individual countries – with different degrees of success – are working to regain credibility and address their debt-sustainability but what is needed is an explicit collective commitment towards fiscal union. In the absence of a clear commitment towards fiscal union and a large enough firewall as a bridge towards that ambitious destination from both monetary and fiscal authorities, investors will not be convinced that Greece is a unique and exceptional case.
The relative calm we are enjoying now may not last for long. Catalysts for uncertainty may only be a few weeks away with the elections in Greece, France and a referendum in Ireland looming. Foreign investors who preferred to stay on the sidelines during the rally may reappear as sellers given the current price levels.
The next few weeks could be a unique window of opportunity for policymakers to shout with a single voice
“all for one and one for all (ex-Greece)” and move in the direction of fiscal union by building up a large enough firewall. The European citizen in me hopes that the next months will not be wasted but the rational investor in me is afraid that policymakers may waste this opportunity, given their penchant for going back and forth between crisis and complacency. Given this view, we think it makes sense to remain cautious on European sovereign risk.