Asia-Pacific Investment Report 2013 — The Guarantee Nobody Wrote Down

2013 delivered two stress episodes in the region from opposite directions, and both were about funding rather than assets.

The external episode followed the developed-world communication described in the Global Investment Outlook 2013. Capital that had arrived since 2009 reversed, and the reversal discriminated sharply: economies running current account deficits funded by portfolio flows were hit hard, while surplus economies with large reserves were largely unaffected. The discriminating variable was funding structure, and it was published and public well in advance.

The domestic episode concerned a class of investment products that had grown rapidly to meet a demand that policy had created.

The structure of those products:

• Investors bought a short-dated product — often a few months — offering a return meaningfully above the capped deposit rate.
• The proceeds funded longer-dated assets, frequently loans to borrowers who could not access or afford regulated bank credit.
• The product was sold through banks, often in branches, to depositors seeking a better return.
• It was legally not a deposit, carried no insurance, and the issuing entity was frequently off the bank's balance sheet.
• And it was widely believed to be safe anyway, because the bank's name was on it and because failures had been prevented before.

Every element of a bank is present here except the protections. Short-term funding, long-term assets, a maturity mismatch, and a run risk — with no deposit insurance, no capital requirement against the assets, and no explicit lender of last resort.

What held it together was an implicit guarantee, and the report's central point is what that does to prices.

What an implicit guarantee does to a price

The mechanism is precise and its consequences are large.

An explicit guarantee — deposit insurance, a sovereign guarantee — is documented, priced, capitalised and disclosed. Everyone knows it exists, who provides it, and what it covers.

An implicit guarantee is a belief that a party will step in, based on precedent, reputational incentive or perceived obligation. It is not documented and nobody has agreed to it.

The pricing consequence is the problem:

• Investors price the product as though the guarantee were real, because from their perspective the expected outcome is the same.
• So the yield does not reflect the underlying asset risk. A product funding risky loans yields only slightly more than a deposit.
• Which means the price carries no information about the assets — the signal that would normally discipline lending is switched off.
• And capital flows toward the guaranteed activity, since it offers a better return for what is perceived as similar risk.

An implicit guarantee removes the risk premium without removing the risk. The gap between the two is real, unfunded, and held by whoever is assumed to be standing behind it.

Why the guarantor tolerates it: denying the guarantee is costly and confirming it is worse. An authority stating clearly that these products are not protected risks triggering the withdrawal it fears. Stating they are protected creates an unfunded liability and worse incentives. So the ambiguity persists, which suits everyone in the short run.

How these situations resolve is instructive and consistent: a small failure is permitted, to establish that the guarantee is not universal. This is genuinely difficult to calibrate — too small and nobody updates; too large and it triggers the general run. The subsequent years contain several such controlled demonstrations.

The Global Investment Outlook 2012 covers the opposite case — an explicit, credible, unlimited guarantee that eliminated a bad equilibrium precisely because everyone understood it.

A bank without the bank parts

The structural comparison deserves setting out directly, because it clarifies what regulation is actually for.

Function	Regulated bank	The product structure
Funding	Deposits, short-dated	Products, short-dated
Assets	Loans, long-dated	Loans, long-dated
Maturity mismatch	Present	Present
Capital against losses	Required, measured	Minimal or none
Deposit insurance	Explicit	None
Lender of last resort	Explicit	None
Supervision of asset quality	Yes	Limited
Loss absorption	Bank capital	Ambiguous

The mismatch is the same. The protections are absent.

Why the structure grew anyway:

• Regulated lending was constrained by loan quotas, sector restrictions and capital requirements. Demand for credit exceeded the permitted supply.
• Depositors wanted better returns than the capped rate, per the Asia-Pacific Investment Report 2012.
• Banks wanted the fee income and the customer relationship without the balance sheet cost.
• And borrowers excluded from regulated credit — smaller private firms, property developers, local entities — would pay materially more.

Every party had a good reason. The aggregate was an unsupervised banking system.

This is the constraint-produces-workaround pattern from the Asia-Pacific Investment Report 2009, appearing at larger scale. A binding restriction on a determined actor relocates the activity rather than preventing it, and the relocated version has fewer protections than the original.

The policy dilemma is genuine. Bringing the activity into the regulated system requires acknowledging it, which implies the guarantee. Leaving it outside leaves it unsupervised and growing. The approach eventually taken — gradual consolidation onto balance sheets with capital required — is slow by necessity, because doing it quickly would force exactly the contraction the authorities were trying to avoid.

When rollover becomes the risk

The interbank stress in mid-2013 is worth examining because the mechanism is the archive's most repeated one, arriving in a new setting.

The structure creating the exposure: products with three-month maturities funding assets with multi-year maturities must be refinanced constantly. Each maturity is a moment when the funding must be replaced.

In normal conditions this is routine and the rollover is close to automatic.

The risk is that it is not automatic, and it fails for reasons unrelated to the underlying assets:

• A general tightening of interbank liquidity raises the cost or availability of funding for everyone.
• A single failure elsewhere causes lenders to reassess similar structures generally.
• Or an authority deliberately restricts liquidity to discourage the growth of the activity — which is what appears to have happened, and which produced a sharp spike in interbank rates.

The consequences when rollover fails:

• The entity must sell assets to repay maturing products. The assets are illiquid loans, which cannot be sold quickly at a fair price.
• Or it must find funding at a much higher rate, which destroys the economics.
• Or it defaults, which tests the implicit guarantee.

This is the funding run from the Global Investment Outlook 2008, in a different jurisdiction, in a different decade, in a different institutional form. The identical structure: short-term funding, long-term illiquid assets, and a lender base that can decline to renew.

Every episode in this archive that involves a maturity mismatch resolves the same way. The assets are almost never the problem. The refinancing is.

The observable early indicator is the interbank rate. A spike means institutions are struggling to fund themselves, which precedes any asset problem becoming visible. It is published daily and free, which makes it one of the better-value monitoring inputs available.

The reversal discriminated

The external episode's most useful feature is how cleanly it separated exposed economies from unexposed ones.

The economies that came under severe pressure shared a structure:

• Current account deficits, requiring continuous foreign capital.
• Deficits funded by portfolio flows rather than direct investment — capital that can leave in days.
• Rising foreign participation in local currency bond markets, which had been welcomed as market development and constituted a large, mobile position.
• And limited reserve buffers relative to short-term external obligations.

The economies that were largely unaffected shared the opposite:

• Current account surpluses, so no continuous external funding requirement.
• Large reserves relative to obligations — the insurance described in the Asia-Pacific Investment Report 2008.
• And lower foreign currency and foreign-held debt.

Every one of these variables was published in advance, by the IMF, the BIS and the World Bank, free of charge, quarterly.

The trigger was unforecastable; the vulnerability was fully knowable. This is the same finding as the Global Investment Outlook 2013 and it is the archive's most actionable recurring conclusion, because it converts an impossible task — predicting events — into a tractable one — measuring exposure.

One qualification worth stating: foreign participation in local currency debt markets is genuinely good for those markets. It deepens them, lowers borrowing costs and reduces reliance on foreign currency debt — which was the great vulnerability of 1997. The 2013 experience does not argue against it; it argues for measuring the position and holding reserves against it.

Products designed by policy

A short structural observation that ties the domestic episode to the previous report.

The product's core appeal was a return above the deposit rate. The deposit rate was capped below market by policy. So the product's entire commercial rationale was a regulatory artefact.

The chain is direct:

1. Deposit rates are capped to subsidise borrowers, per the Asia-Pacific Investment Report 2012.
2. Households receive poor real returns and seek alternatives.
3. Institutions supply an alternative yielding more, funded by lending to borrowers excluded from regulated credit.
4. The alternative reproduces bank risk without bank protections.
5. Authorities must now supervise a system created by their own price control.

The generalisable principle:

A price control creates demand for whatever evades it. The evasion is usually less transparent and less protected than the thing controlled, and the controlling authority inherits responsibility for it.

This is not an argument that the price control was wrong — it funded a growth model that delivered enormous real gains. It is an argument that the second-order consequence was predictable from the policy, and that the eventual liberalisation of deposit rates was as much a financial stability measure as a rebalancing one.

How to detect a guarantee nobody has given

Since the pricing distortion is the investable consequence, the practical question is how to identify an implicit guarantee before it is tested. Five diagnostics, all usable with public information.

The yield gap is too small for the assets. This is the primary signal. Compare the product's yield to what a standalone lender funding the same assets would have to pay. If a vehicle lending to property developers yields barely more than a government-guaranteed deposit, something other than the assets is supporting the price.

Marketing and distribution imply a relationship the documents deny. Products sold in bank branches, by bank staff, under names resembling the bank's, while the legal documentation carefully establishes that the bank is not liable, is the clearest available tell. The gap between how something is sold and how it is papered is where the implicit guarantee lives.

Past failures were prevented rather than resolved. Look for precedent. If previous stress in similar products ended with investors made whole through an arrangement nobody was obliged to make, the expectation is established — and each such rescue makes the next one more expected and more expensive.

The guarantor has never denied it. Authorities facing an implicit guarantee they do not want typically say nothing rather than deny it, because a denial risks triggering the run. Persistent official silence on whether a product is protected is informative, and it is observable in what regulators publish and do not publish.

And the investor base cannot bear the loss. A product sold to retail savers in small denominations carries a political obligation that a product sold to institutions does not. The likelihood of intervention is a function of who holds it, which is disclosed in product documentation and regulatory filings.

An implicit guarantee is not hidden. It is visible in the gap between the yield and the assets, and in the gap between how the product is sold and how it is documented. Both gaps are measurable from public information.

The investment consequence cuts both ways, which is what makes this practically useful rather than merely cautionary. Where a guarantee is likely to be honoured, the product may genuinely be a good risk-adjusted holding — the investor is being paid a spread for a risk that a third party is bearing. Where it is likely to be tested, the yield is far too low. The diagnostics above are how you form a view on which situation you are in, and the answer determines the direction of the trade rather than merely whether to avoid it.

What an allocator could act on

Look for the risk premium that should be there and is not. A product funding risky assets at close to a deposit yield is priced on an implicit guarantee, and the gap between yield and risk is held by whoever is assumed to be standing behind it.

Identify maturity mismatches outside the regulated perimeter. Short-dated funding against long-dated illiquid assets is a bank, and the absence of insurance and a lender of last resort makes it a more fragile one.

Monitor interbank rates as a leading indicator. They are free and daily, and a spike signals funding stress before any asset problem becomes visible.

Screen external vulnerability by funding composition, not deficit size. A deficit funded by direct investment and one funded by portfolio flows carry entirely different reversal risks, and the aggregate figure conceals the difference.

Measure foreign holdings of local currency debt. Local currency issuance removes the currency mismatch and creates a large mobile position, so the improvement and the exposure arrive together.

Ask what a price control creates demand for. The evasion is predictable from the policy and is generally less transparent than what it replaces.

What 2013 established

• Implicit guarantees remove the risk premium without removing the risk, switching off the price signal that disciplines lending.
• Maturity transformation outside the regulated perimeter reproduces bank fragility without deposit insurance or a lender of last resort.
• The products were created by the deposit rate cap, making liberalisation a stability measure as well as a rebalancing one.
• The external reversal discriminated by funding structure, which was published and knowable in advance.
• Interbank rates are the observable early symptom of a rollover problem.