Tuesday, March 3, 2009

The Weakness of the Ringgit & the Finance Minister

Why is the ringgit so weak? Is this part of policy strategy?

I read with bewilderment the comment by FMII in Parliament that Malaysia is not in recession - does this mean that the government is in denial?

To argue the recession on technical ground - of contraction in two consecutive quarters (how do you measure that?) - is a great debating point but an insult to intelligence.

Why is the economy not in recession?

1. The government does not know how to measure quarterly GDP. Do you do a year-on-year growth for the quarter to get the change? Or do you do the real thing by first deseasonalising the quarterly GDP and calculate the change against the preceding quarter?

2. Is Malaysia different from the US? If it is, then the technical definition of a recession may not be relevant - a definition created by the US National Bureau of Economic Research. What should be Malaysia's definition of recession? Growth below 3%?

3. Malaysia has already been in recession for too long already. Maybe because Malaysia has been in recession since the end of the Era of the Mega Projects which left a hole in the economic structure and psyche - and our Series of Little Doses of Government Spending is our way of bypassing that psychology but still engaging in the stimulation. If the economy is already in recession, you cannot drop any further - you are already at the bottom.

Why is the ringgit so weak across the board?

1. Interest rate cuts. Because of the cuts in the interest rate which means that savers are going to other currencies to learn higher interest rates. But all interest rates have been cut to near-zero except for Australia and New Zealand.

2. Loss of Investor Confidence. The cuts in the interest rate may be an indication that the economy is really in trouble and the central bank is merely protecting itself from criticism by "pre-ampting" the recession which may be coming. With no hope in listed companies - which are looking for cash from investors to bail them out - investors are fleeing from stocks in the market, selling them and getting out of Malaysia.

3. Anti-Deflation Policy. Maybe the policymakers are trying to counter the possibility of deflation in Malaysia by creating inflationary pressures from a weak currency. If so, this is a very sophisticated policy move.

4. No Support for Ringgit. The authorities have decided not to support the currency in order to preserve their holdings of foreign reserves - as well as to inflate the ringgit value of the foreign reserves which, by definition, are all held in foreign currencies. The authorities have boasted about the strength of the local economy by the amount of foreign reserves that we have. Has this been a tactical error?

What for the stimulus package?

1. If the economy is not in recession, then there is no need for another stimulus package. Or, what was meant is that the economy has not been in recession in the last year - but that it could be in recession in the new year. If this is the case, then there is deliberate attempt at being pedantic. Policymakers are looking at history not the future.

2. We want to know what is the problem with the economy as a starting point for policy. The bloggers seem to know. But the bloggers want to be reassured that the authorities know. We live the Great Age of Globalisation. We are at risk of knowing what is happening around the world, but not the problems in our own little backyard.

3. We should create opportunities for the young generation to venture into business - all members of the young generation and not just a select few. If we cannot interact among ourselves at home, how are we going to be able to interact with the rest of the world. Create opportunities by providing means of interaction and exchange of ideas - with the financiers (microcredit or otherwise) always at the background ready to seize the opportunity to lend. Or, has our system been spoiled for mega projects such that financing the little men and women has become not a business proposition but a tedium?


hishamh said...

A few points I'd like to take up here:

1. First, re: calculation of GDP growth. 2008:04 growth based on quarter on last year's quarter is as you know 0.0%. Based on seasonlly adjusted q-o-q, annualised, rGDP shrank 10.1% in Q4. I suspect even that will eventually be revised downwards, when the full trade numbers are in.

2. The NBER definition of recession is not 2 quarters of negative growth (which is the way I read your comment on this). If it was, the start date of the recession in the US would be in Q4 2008, rather than Dec 2007. The 2 quarter definition is just a generally accepted rule of thumb, and not really an official definition by any organisation I know of.

3. I would dispute the characterisation of the Ringgit as being 'weak across the board'. On a trade weighted basis, the Ringgit is down only 1% in nominal terms since Dec 2008. The 'real' picture might be different, but I'm lacking enough data on inflation rates in China to make a good determination at this point. Given the short lag, however, relative movements should be approximate. Major currency movements relative to the Ringgit over the last two months have been in USD, CNY, HKD and VND (-7%), GBP (-5%), and KRW (+12%).

4. Regarding your point 2, my hypothesis is that global deleveraging and flight to cash have been primarily responsible for the withdrawal of portfolio funds from Malaysia (and many other emerging markets), not an investor loss of confidence. To put the numbers in perspective, a net RM56 billion in portfolio investment left the country in Q3 2008 (BOP data), and the process accelerated in Q4 (my back-of-the-envelope estimate). In other words, funds have been leaving the country before the economic data started turning down.

On the whole, I'm pretty happy with the way BNM handled the sudden loss of liquidity - growth in monetary aggregates is down, but not to the point of making monetary policy tight. They're running out of balance sheet bullets, though.

etheorist said...

Dear hishamh,

Thank you for your comments.

I support your "econometric" approach to your economics blog. I have put it up in my links. Welcome and contribute.

Of course, you know I do not quote nunmbers in my pieces. I look at trends, feel them and then think about strategic responses. I integrate my thoughts between theory and the real world as I understand them. So your blog will be a good complement to what I do.

My response:

1. With GDP down sharply and since the decline is not seasonable, we do not expect any appreciable pick-up. Yes, the worse is yet to come.

2. I think NBER has changed their minds and become more sophisticated with their definition. The proper definition of a recession is from the top of the cycle to the bottom which of course dates it earlier. But it is not intuitive to the public. So when there was a contraction in the economy, the wisdom was that one swallow does not make a summer. So the convention wisdom was to expand it to two quarters. Somebody attributed this rule of thumb to the NBER long ago - if my memory serves me right. Thanks for the technical correction.

3. I look at the current ringgit exchange rates and it has weakened against all major currencies - even against the Thai baht - from the last we all noticed when the world currencies have weakened and the ringgit has some semblance of strength. That was when the world was cutting interest rates, and this is when Bank Negara cut interest rates further. If the ringgit basket falls even by 1% nominal, that is decline across the board. I think there has been a serious decline of the ringgit across the board and I am not happy because I do not understand why this should be the policy stance. We all know the USD has strengthened against everybody else - but why is everybody else strengthening against us.

4. There is an even more subtle point I am trying to make. Of course, the world is deleveraging. US companies are pulling cash back to HQ. US is raising more bonds to finance their deficit. But why this sudden weakening of the ringgit - after the further cut in the OPR for no apparent reason. Or is the reason that the central bank is expecting the economy to be really bad in the coming months and year and the stock market knows it.

I enjoy your comments and do come and visit us often.

hishamh said...

Thanks for the quick reply!

Re: your comments regarding the Ringgit. Yes, the Ringgit has been generally depreciating for most of 2008, and not just against the USD. The overall movement amounts to 1.8% in nominal terms, and 2.9% if you include the first two months of this year.

I don't see this as necessarily being a bad policy on the part of the authorities. Assuming a 2.5% trading band around the target rate, a 3.0% trade weighted movement would be excessive only if you presume the Ringgit was fairly valued at the beginning of the decline. My long term model of the Ringgit suggests (insert disclaimer on model specification errors, etc.) that the currency was overvalued by as much as 5% in real, trade-weighted terms at the end of 2007. That puts a more benign interpretation of Ringgit movements since then.

etheorist said...

Dear hishamh,

I am glad to hear you have an econometric model on the currency because most of us do not.

My two comments on currency models:

1. The reference point is important. Which year is equilibrium. This brings out sore points for oldtimers who, for example, cherish 1980 being a golden year. This is a very hard issue to resolve me.

2. For real exchange rate, which is the better indicator of the price level. Using the CPI understates inflation in Malaysia and generally gives the result of an appreciation in real terms. But I think the model should use asset prices (both financial and real) as the price indicator because this is where most of the money supply gets translated.

Keep working on the models.

hishamh said...

1. I'm not sure what you mean by reference year - are you thinking in terms of a PPP or macro balance framework? I'm using 2000=100 as the base year for the real exchange rate index (sample:1972-2007), but using different base years has no impact on estimating the VAR or VECM of the real effective exchange rate (apart from the intercept(s)). In this instance, I'm using a reduced form equilibrium model in a Johansen cointegration framework - the same methodology is commonly used by the IMF. The advantage of this approach is it's atheoretic, and does not require a-priori normative estimates of equilibrium.

2. Quite remarkably (at least I found it so), in practice the price deflator doesn't really matter - the calculated indices are so close, they're virtually equivalent. More important is choice of curencies, the weighting scheme, and frequency of weight changes.

etheorist said...

Yes and no.

1. Indexing is neutral for model estimation. But there is a base for calculating change. I am thinking of long term trends.

2. Even for real exchange rate models, you still have to refer to a base point to say where it has appreciated. My point was that if it is very close even 1997, we may not get the true picture.

3. The problem is that the currencies are much more volatile than prices so you will find a low correlation is the model is underspecified. I would use more variables (government spending or deficit, stock market indicators, trade balance) to take care of the big changes so that relatively small changes can play a role - provided you have sufficient degrees of freedom. Welcome to the art of econometrics.

hishamh said...

Ah, now I understand. The nice thing about the estimation I used is that equilibrium is endogenous to the model. And yes, I know the pitfalls in using such an approach.

The actual variables used are Gov consumption/GDP, PPI(domestic)/PPI(imports) as a proxy for the terms of trade, trade duties/total trade to proxy trade openness, PPI/CPI as a proxy for relative price of tradables to non-tradables (Balassa-Samuelson effects), Net Foreign Assets/GDP, and the real interest rate. All of these except openness and BS proxies are relative to trade partners. There's also a dummy variable to account for a potential structural break in 1997-98.

As you can imagine, conserving degrees of freedom was a big problem. The final specification I ended up with however had coefficients that were all significant and with the correct signs, except for the real interest rate (which had neither).

VECMs as you may know can be decomposed into long term trends and short term movements. The long term component in this formulation is interpreted as the 'fundamental' equilibrium exchange rate, which would then be the baseline for determining over/under valuation or equivalently appreciation or depreciation relative to equilibrium.

etheorist said...

Very good.

satD said...

hi etheorist n hishamh..

Fuh banyak takut mau kasi komen now..

i got lost half way in between....

never believed in econometrics models on currencies....

If you have flow data on real time basis then maybe boleh pakai for a market that moves that fast...24 hours a day...

Real time position reporting would allow for that to be monitored..

Big problem.....OTC market..massive infra to build to achieve that ..but if all central banks do it in their domestic market n share this flow info privately purely for internal use....

satD said...


yes..multi-generational policies now

hishamh said...

satD:The model I'm using basically tries to measure the degree of misalignment between the "real effective" value of the Ringgit (a synthetic construct) and the same value consistent with economic fundamentals - you can get an idea of whether the currency is under or overvalued relative to the currencies of all trade partners, as well as an idea of how long it will take to move towards equilibrium. This makes it a good guide for policy and potential central bank intervention.

However, this model is not useful for doing forecasts, unless you have a time horizon of over two years (after which of course the fundamentals might change, and you get a different target!). For day to day trading the best model remains the "random walk" model, which means you can forecast volatility pretty accurately, but not magnitude or direction of the market.

satD said...


Tks for ur reply.

As for magnitude n direction if anyone can do then kaye raya by now.

So far hows d performance of ur model?

hishamh said...

I don't think 'performance' is the right word for the model I'm using - as I said it's not a forecast/prediction model, so there's nothing to evaluate performance against.

Anonymous said...

Somebody told me a simple formula. Recession is when your neighbor lose his job. Depression is when you lose yours.

etheorist said...

satD and hishamh,

Econometrics models are great and we should play with them a lot more - in order to understand that little of the future is like the past.

I like what hishamh is doing for it will put people like me on the straight and narrow path of mathematic logical based on the historic.

I am now mindful that I have to get my "facts" right.

However, the challenge for the future is to predict based on a cosmic view of the economic world.

We have to increasingly advance our thinking on the nature of our world of which the economics is a small but important part.

I like this exchange.

hishamh said...

etheorist: I think you've pegged me wrong. I don't blame you for it, as relying on statistical arguments generally implies a belief that the future can be predicted by the past.

However, despite my focus on econometrics, I almost never issue or work out forecasts (especially point forecasts), except in very general terms or when I want to show that it's wrong - despite the fact that I work as a professional economist in the financial industry.

I know enough about the weaknesses of modelling not to go along with the rest of the industry on that score.

My interest in econometrics (and one of the reasons why I decided to take up blogging) is primarily with a view to finding out where we are now, as a guide to where we want to be and how to get there. There is a distressing tendency in the media, the public, and the economics profession to blanket apply the theoretical constructs of economics to the real economy, without questioning the underlying assumptions and reality.

I am as guilty of that as anybody else, in the areas that are not my expertise.

What I'm trying to do within my limited scope of knowledge, is work the other way - what are the numbers saying relative to economic theory and what people believe is going on? That is really the core of econometrics, not forecasting.

It is also what prompted me to comment on this post - to me, the Ringgit depreciation story is really a non-story, when viewed from the perspective of trade.

I'm fully in agreement with you that we need to understand our world and our people better to progress. But initial conditions matter, as does the experience of others in the past. As the philosopher Georges Santayana once said, "Those who cannot remember the past are condemned to repeat it." There is a value in confirming or disproving the prevailing 'wisdom' - it provides a frame of reference for future discourse. Otherwise, any debate on policy becomes just a clash of ideologies, lacking in substance.

Your comments on my posts help remind me that in all cases, different perspectives and interpretations have validity - in other words, you're keeping me on the 'straight and narrow' as much as you say I have done for you.

So I hope we can continue this exchange - I've been enjoying it immensely, and I truly value your input.

etheorist said...


I am happy with what you are doing and I fully support it.

My comments are with respect to the subject in general, and I thought I can provide some insights.

I am sorry if it appears personal which it wasn't intended.

Keep on working and exchange ideas.