Despite a worsening inflation-growth tradeoff in Vietnam, there are also some factors that could help counterbalance the risk of an imminent BOP crisis such as the kind Thailand suffered from in 1997.

1) FDI strength likely to continue and portfolio investment still trickling in

Vietnam has been and will likely continue to finance its trade deficit with FDI inflows, official development assistance (ODA), and private remittances. While the trade deficit widened in the first five months of this year, FDI inflows have more than doubled from the level of a year ago to US$14.7 billion with healthy disbursement inflows. We also expect forthcoming FDI inflows to hold up in the rest of 2008, considering concomitant capital in large projects initiated in the recent past and a stable flow of investment in the petroleum-related industries and IT manufacturing. Anecdotal evidence also suggests remittance inflows from overseas Viet Kiaus have remained strong.

In addition to steady FDI inflows, portfolio investments coming into Vietnam have remained positive, and those who chose to stay until now seem unlikely to leave unless they lose confidence over the medium-term growth outlook. During the sharp decline of 56% in Vietnam’s equity index since the beginning of this year, foreign investors have been net buyers (while domestic retail investors were the sellers) throughout the period. Still, even during the past two days of HCMCSE suspension, foreign investors remained net buyers in the Hanoi stock exchange (albeit at a much smaller position).

2) Short-term overseas borrowing is limited

In contrast to Thailand in its pre-Asian crisis period, Vietnam has a much smaller position in overseas short-term borrowing and has never relied on it to finance its trade deficits. The World Development Indicators show Vietnam’s short-term debt outstanding as only approximately 8.6% of GDP in 2006, in contrast to Thailand’s 26.3% in 1996. Given the central bank’s tight regulations on overseas borrowing, Vietnam’s short-term loans have been small in size and mostly related to trade credit, and there is little evidence that this has changed dramatically in the past 12–18 months. In addition, Vietnam’s total external debt (US$22 billion) is slightly larger than its official FX reserves in size, but most of it is medium-to-long term debt is on highly concessional terms.

3) The increase in onshore USD demand has mainly come from imports, so far

The unusual jump in imports growth driven by stronger investment demand in recent months will likely normalize soon on the back of the monetary tightening through credit control, especially at a time when ports are running out of capacity for further import storage. Meanwhile, although it is possible that some imports over-invoicing could have been used to disguise capital flight, our discussions with local business contacts on the ground have offered little support of this being a widespread practice.

In the offshore market, VND NDF rates have been pushed up by traders unwinding long positions in the VND and adding positions to hedge against currency risk in VND sovereign/commercial bond holdings. Despite the large volatilities and pessimism priced in the NDF prices, its dynamics have remained relatively isolated from onshore USD demand so far.

With the existing FX controls under the capital account in Vietnam, the likelihood of an outright currency attack on the VND initiated by international speculators remains limited, in our view. However, if inflation deteriorates further for a sustained period of time, local capital might flee into gold and the USD, putting the domestic monetary system under stress.

To reduce inflationary pressures, we believe the Vietnamese government will likely have to extend price controls after they expire in June and undertake more austere fiscal measures by cutting expenditure and public investment projects to stabilize the economy. The current plan of cutting public spending by D2-4 trillion (compared to D112 trillion of government spending last year) does not seem to be sufficient to curb inflationary pressures effectively. In addition, the central bank will likely tighten credit controls, especially after commercial banks attracted more deposits after the deposit rate ceiling was removed two weeks ago. Although these measures should help contain inflationary pressures, they may tend to weigh on domestic equity prices (especially financial and property names) in the near term.

On the currency front, our current assessment is that the probability of the central bank being forced into taking an abrupt and sharp nominal devaluation in the near term is still low. In addition, given the potential impact of a large currency devaluation on inflation, the State Bank of Vietnam (SBV) will unlikely propose this option actively, especially when it is hard to determine how much a one-time adjustment in currency value will be needed to balance the mismatch in FX asset/liability positions.

However, we believe it has become more likely for the monetary authority to consider the option of an accelerated devaluation against the USD, to prevent the VND from being excessively overvalued in real terms. In the first four months of this year, although the VND depreciated by 4.1% in nominal terms against its major trading partners, it strengthened by more than 11% in real terms due to its higher inflation level relative to that of other countries. Since we expect CPI inflation (especially in yoy terms) to stay at elevated levels for the next few months, the SBV could risk pushing the VND official exchange rate to its highest level in real terms since 1997 in 4–6 months if it chooses to stick to the current 2%-per-year pace in VND depreciation against the USD in official parity. By then, risks on a currency crisis triggered by BOP difficulties may escalate to a less manageable level, and remedies could also be more painful.

The weakest part of the debt chain is usually a good place to look for marginal stress in the system. Pawn brokers have increased monthly lending rates from 4% to 6% in recent months and reduced loan/value ratios to 70% for gold and 50% for diamonds and motorbikes (my wife still refuses to believe me that diamonds are a rubbish asset class ). BUT, default rates have not yet shot up – there is some terming out of 1 month loans into 2/3 months, but the impression is that borrowers are still hoping conditions will improve soon. This was before the recent lifting of the deposit rate cap, which will obviously lead to even more stress in the curb market, as funding costs escalate along the financial chain.

12 month forward premium for Vietnam Dong

The offshore “Dong carry trade” is being unwound, in an increasingly desperate fashion. Around late 2006, foreign bank prop traders and hedge funds started to pile into VND bonds, typically with up to 3 year maturity, at around 7-8% yield. The supposed asymmetric risk (for a VND revaluation) was viewed as outweighing the lack of liquidity and hedging tools. Market estimates of the size of these inflows are USD5-7bn. The problem was that these carry traders indirectly ended up funding a property bubble, as the SBV failed to sterilize. Now, they (or their risk officers) want out and there’s no market. Bonds have sold off to 15%, and the USD/VND NDF market has been trading at distressed levels, with the 1-year USD/VND offered around 19,000, implying a 15% devaluation from the current SBV ceiling rate (see Graph 1). Market estimates are that about one quarter of the positions have exited, implying a big overhang on VND/USD.

There is another source of potential BOP risk, namely the local hedging out of short USD/VND risk. Onshore USD loans via the banking system are estimated at about USD13 billion. This has accelerated over the last year as locals put on the carry trade as well, explicitly or implicitly by shifting their VND funding into USD (so-called “liability dollarization”). This also showed up as a sharp fall in the net foreign assets of the banking system, as onshore banks drew down their offshore US dollar lines to fund onshore USD lending. This provides another source of credit risk for the banks (as the VND devalues).

The VND is weakening, despite SBV intervention. Obviously, because of the above factors, VND/USD has shifted from the strong to the weak side of the SBV band. They have been selling USD to defend the band since end-March, and the market estimates are around USD2 billion (which would imply official reserves still around USD21 billion). However, the SBV has not been prepared to “fully” defend the ceiling rate, which has led to the development of a shadow price for USD/VND. The clearing level for USD/VND is derived from where EUR/VND (which is not subject to the same legal restrictions) is traded. The actual implied USD/VND has recently traded around 16500 or 2% outside the band (see Graph 2).

The proximate cause was a major macroeconomic management error. The underlying cause was a sequencing (of reforms) error – a fact that has been curiously absent from all the breathlessly bullish analysis of the Vietnam story.

So, what went wrong? As numerous people have pointed out over the last few years (nice timing by The Economist a few weeks ago!), the Vietnam structural reform story has been a great success since the end of the Cold war (I still remember visiting Ho Chi Minh for the first time in 1991, and everyone assuming I was Russian!) And, none of the medium term positives have really changed.

But, even high growth success stories like Vietnam have a growth/inflation trade-off. After China opened up in the late 1970’s, it had a series of mini-boom/bust business cycles culminating in hyper-inflation and a mini-BOP crisis by 1993. Beijing then realised that it had to make a commitment to building a credible central bank (amongst other important institutional reforms) and spent the next seven years under Zhu Rongji constructing that platform for more sustainable growth. And THEN they were ready to manage the positive BOP shockwaves from WTO entry.

Vietnam’s structural mistake was to take a “convoy” approach to institutional reform. They wanted to reform State Owned Commerical Banks (SOCB’s), SOE’s, the SBV, bank regulation etc. all at the same time, using the pre-commitment device of WTO entry. This was a major sequencing error. They should have done the hard work on institution (and infrastructure) building before going so aggressively for short-term growth.

This was the background for why the torrid boom of 2005-2007 is now turning to bust. The SBV lacked the tools, mandate, independence and manpower to deal with a positive BOP shock. Its FX reserves and balance sheet trebled in two years, and they only partially sterilized allowing the additional liquidity to fund a blowout in the largely unregulated commercial banking system. Meanwhile, fiscal policy was also on an expansionary track as SOE’s sought to accelerate infrastructure building, funded via the state-owned commercial banks. This hyper loosening of the monetary and fiscal policy mix was a major macro management error.

Why did nobody warn of these obvious risks? Good question. To the IMF’s credit, they were the only institution raising a red flag on potential inflation risks as long ago as 2006. But it seems like their voice in Hanoi (office size 6 people) was somewhat drowned out by all the other breathless admiration from the NGO’s, supra-nationals and investment banks. I think Vietnam is still the largest single recipient of ODA (official development assistance) – which might have something to do with it! Maybe Hanoi drunk the cool-aid about being the next China, and believed the hype/flattery.

The policy response will be the critical variable determining the scale of the crisis. This is where China’s austerity program under Zhu Rongji in 1993 provides a helpful template.

So, what’s the plan from here? It is clear from recent interviews that the Prime Minister recognises that an inflation problem exists, stemming largely from a mistake in monetary policy. However, there is a reluctance to recognise that short-term growth targets may have to be sacrificed. It is important to remember that this is the first financially-leveraged downturn that the economy has ever experienced. There are no historical precedents internally to reference from. The response so far has been quite fragmented, limited to stop/go liquidity controls and fire-fighting the consequences – maybe trying to emulate the PBOC’s response over the last few years. China’s austerity program in 1993 would be a far more appropriate policy template to study.

The SBV tightened liquidity conditions sharply after Tet (early February). Reserve money contracted 19% in March alone, having exploded by 27% in the first 2 months of the year on end-2007 levels! They seem to have reopened the liquidity tap in the past few weeks. The extreme volatility in overnight interest rates in the last six months suggests a central bank that is panicking. Adding to this sense, is the fact that commercial banks have shifted from basically having no regulatory prudential control to a system of having to report their balance sheet daily to the SBV! The use of arbitrary deposit rate ceilings also has added to the confusion on the funding side; if interest rates are not allowed to clear, then nobody should be surprised if banks get caught short.

The results of this uncertainty are predictable. Banks are afraid to lend and are desperately trying to eliminate their excessively high exposures to the asset markets. Hence, the daily limit down in the equity market, on minimal volume. Letters of credit are getting harder to secure, domestic trade credit terms are shortening, and the whole payments system is under pressure. Welcome to the world of financial accelerators.

It was interesting to see how the recent liquidity injection translated almost immediately into a weakening in the unofficial market rate for VND/USD, and failed to provide any relief to the equity market. Clearly, the situation has deteriorated to a position where a broader policy response is urgently needed.

Unfortunately, the scale and quality of the Vietnamese credit boom over the last few years makes Thailand 1994-1997 seem like a tea party.

So, how serious is the Vietnamese crisis on a regional scale? We have seen this sort of credit boom/bust before in Asia many times. Is this one a more benign example like Indonesia 1991-1993 (following the Pakto big bang financial reforms of Oct 1988) or Thailand 1995-1997? Unfortunately, this one looks serious, and maybe very serious depending on how well managed the policy response is.

What distinguishes the Vietnam case is pace, scale and the quality of the credit boom. Over a 15 month period (end-2006 to March 2008), an unregulated banking system increased its loan/GDP ratio by about 30% points (or USD30 billion), mainly into asset-related areas. A society that likes to gamble started to access leverage on an unheard of scale for the first time. Graph 3 shows how big an outlier Vietnam became in the last few years, by scaling credit/GDP to per capital income on a regional comparison The strange thing is that if you view Ho Chi Minh from the top of a tall building, there is a surprising lack of cranes at work. This makes me suspect that banks became property traders themselves in this cycle!

As the monetary survey overleaf illustrates (see Graph 4), bank credit/GDP is now 100%, which is a major outlier when scaled against per capita income. Looking at a comparison of Thailand 1994-1997 helps provide some context. The increase in the credit/GDP ratio (to the same 100% by 1997) was 20%, over a four year period, in a much more sophisticated financial system and in a country with four times the per capita income.

This is why a short, sharp monetary shock will not solve the problem on its own. The system is so top-heavy with debt now that even a small tightening has led to a chain reaction of debt deflation almost immediately. Sure, inflation slows, but at an unacceptable political cost to growth.

An “IMF program” style policy response will be needed within six months. This should involve a sharp tightening of monetary and fiscal policy, a sizeable VND devaluation, the effective nationalization of most of the commercial banking system, and the establishment of a “bad bank” to facilitate the necessary large scale NPL carve outs.

The good news is that Vietnam has a strong government, with a good track record of making the hard decisions when it matters. This increases the chances of Vietnam going the “China 1993” route i.e. a home-grown inflation stabilisation programme. The challenge will be for policy makers to compress about two years of institutional reform and education into the next six months. But a lot of political capital will need to be spent fast, given the size of the problems.

The response will need to involve both a unified macro response (involving monetary, fiscal and foreign exchange policy) AND a financial sector reform response concurrently. Recapitalising the banking system efficiently will be a key determinant of the pace of recovery from the coming “growth recession”. A sizable VND devaluation (into a more flexible managed float, probably) WILL be needed. A competitive export sector will be needed to provide macro oxygen to offset the immediate collapse in domestic demand.

There are likely to be a number of political hurdles to be faced. Hanoi needs to get tough with the SOE’s and provincial government expansion plans, including non-essential project freezes. Who will absorb the weaker banks? The government will probably want the Tier 1 commercial banks to do some “national service”, but resistance will be high. Who will establish and manage the “bad bank”, and establish the transfer price for NPL’s to this facility? Lastly, no government likes to devalue their currency, voluntarily. A stable VND has been viewed as an important competitive anchor – the problem with this argument is that the VND has not been stable in inflation-adjusted or real terms. There will be a concern about second-round effects on bank NPL’s from the unknown amount of unhedged USD borrowings. And there will probably be some sovereignty issues incorrectly mixed into the political debate also.

Whether this will be a home grown policy response or formally imposed by the IMF is unclear at this stage. The latter might end up being the more politically expedient path, if the domestic political issues prove too complicated. The strength of the political response to a crisis is always a key issue. This is the point around which the “China 1993 versus Thailand 1997” debate really revolves around.
The fiscal costs are likely to be manageable, given the underlying growth dynamics of the country. Indeed, this coming crisis could be exactly the shock needed to push through a more serious commitment to institutional reform – and lay the foundations for more sustainable growth profile in the coming decade.

This is not a crisis of the Vietnam economic model. It is a cyclical macroeconomic error that got out of control due to a weak regulatory framework. ASEAN’s mistakes of 1989-1995 (the move from undervalued to overvalued exchange rates) were compressed into two manic years in Vietnam. It just illustrates the critical importance of strong institutions. Vietnam will recover from this, probably a much stronger member of the globalised economy.

Government debt will rise in the short-term but total public debt levels of 43% of GDP leave room for the necessary fiscalisation of the banking system clean up costs. In any case, the majority of the NPLs lie with banks with some form of government ownership. And half of the public debt is foreign, so even a 20% devaluation would only add 4% points of GDP onto the overall debt stock. It is going to be extremely bumpy in the near-term but one should not lose site of the fact that this is a high-growth economy which can easily improve its debt sustainability profile once the economy is restored to health.