This morning, I had the distinct displeasure of handling some small denominations of US dollars here in Harare. Although, I really should have started with “But I’m really glad that I had cash to handle in the first place” – because in the weekend panic about the rumoured early issuance of bond notes (yesterday, apparently), there have been chaotic queues at almost any Zimbabwean banking institution that is still issuing US dollar bank notes.
Some background for the non-Zimbabweans:
- Zimbabwe uses the US dollar.
- “used to use”
- It was adopted by the Zimbabwean government under the multicurrency basket regime that took over from the Zimbabwean dollar in 2009, when hyperinflation had finally driven the local currency into a near-bottomless grave.
- However, from then on, economic fundamentals meant that Zimbabwe was slowly but surely running out of US dollars.
- Until the beginning of the year, when it began to haemorrhage them.
The basic problems:
- Zimbabwe spends more on imports than it receives in exports receipts.
- Until recently, the ‘extra’ needed was being topped up by remittances from the diaspora and some foreign loans.
- But then a few things happened all at once:
- The commodity cycle turned (so Zimbabwe was earning less money for its mining exports);
- Commodity-dependent export services suffered a business slow-down;
- The currencies of the surrounding commodity-dependent countries weakened (so local Zimbabwean manufacturing companies because less competitive);
- The diaspora in those surrounding commodity-dependent countries were sending the same amount of locally-denominated money home to their families in Zimbabwe, but it was worth significantly less in US dollar terms due to their currency weakening;
- The drought meant that any agricultural exports were lower than expected (and the need for food imports was higher than expected).
- The RBZ and the Minister of Finance apparently tried to pay back some of the arrears on the IMF loan (and a few other long-outstanding loans).
- Some erratic government policies meant that foreign loans dried up.
- At this point, the Zimbabwean balance of payments began to look real shaky.
- Then it became clear that all was not as it should be.
- A shaky balance of payments is one thing – but it is an altogether different thing when civil servants are being paid in export proceeds.
Here is a bad Balance of Payments situation:
Export Proceeds + Foreign Loans + Remittances < Import Payments
Export Proceeds + Foreign Loans + Remittances < Import Payments + PUBLIC SECTOR WAGES
To be clear – this is not my own explanation of the problem. This is the Reserve Bank governor’s explanation of the problem – in his September 2016 mid-term monetary policy review.
Here’s the image that was helpfully included in the presentation:
Nostro Accounts: are the foreign bank accounts held by a local bank at a foreign bank (called the ‘correspondent bank’). Basically, the idea is that a US dollar payment from South Africa to Zimbabwe is not actually a US dollar payment from South Africa to Zimbabwe. Instead, the South African bank asks its correspondent bank in New York to move money to the Zimbabwean bank’s account at its own correspondent bank in New York. When the Zimbabwean bank hears that its correspondent bank in New York has received the payment, it then credits the account of the recipient*. Or, more simply, the foreign payment involves the South African bank transferring money from its nostro account to the Zimbabwean bank’s nostro account. And it just so happens that both of those nostro accounts are held by correspondent banks in New York.
*I know, right? No wonder foreign bank transfers take so long.
RTGS Accounts: are the local accounts within a country (or, at least, that’s what the Zimbabwean banking system calls them).
Here’s a picture that may help (remember that each bank will have a ‘nostro account’ at their correspondent bank):
Nostro Accounts: hold money which can be spent anywhere in the world.
RTGS Accounts: hold money which can only be spent in Zimbabwe.
This might sound scary, but all it really means is that your local payments get swapped between local banks, and your foreign payments get swapped between the nostro accounts of local banks at their respective correspondent banks.
And because there’s usually a higher demand for local payments, it’s really fine if those amounts don’t quite equal each other. Especially as the local banks will still engage in a bit of money creation (that is: they’ll extend some loans – which has some fractional banking impacts irrespective of whether the economy wants it or not).
In Zimbabwe’s case, there was a really high demand for cash notes. And because of that, the Reserve Bank of Zimbabwe (the RBZ) was regularly cashing-in its nostro account at the New York Federal Reserve in order to import nice clean USD bank notes back to Zimbabwe.
When you start cashing out your nostro account in order to pay civil servants in ‘real’ currency (especially given that almost the entire tax revenue collection gets spent on civil servant salaries)… Well now you’re effectively spending export proceeds on the public sector wage bill.
To reiterate: the country was collecting local RTGS payments from taxpayers, and converting those into real-cash-nostro-account balances for civil servants.
This is especially not great when your importers are already consuming more of the nostro account balances than the exporters can replenish:
So when all those issues I mentioned up top started to happen, the RBZ inevitably began to run out of money in its own nostro account. It then imposed new rules on the local Zimbabwean banks to deposit even more of their nostro balances into the RBZ’s nostro account, in exchange for local RTGS balances in Zimbabwe.
Those too were cashed in.
Then when local importers came to pay for their imports, the local banks had no money in their nostro accounts, and had to go request some from the RBZ. And then the payment processes started to be delayed until exporters paid money into Zimbabwe – at which point, the RBZ nostro balances would have new money in them, and the local banks could then process the importer payments (if they were high enough up the priority list).
Now we’re a few months down the road from there, and we’re at the point where the RBZ desperately wants to stop depleting those nostro balances in exchange for cash notes – cash notes which then leave the formal banking system permanently because everyone is worried that it might not be there tomorrow.
So the RBZ is proposing bond notes, backed by a foreign credit facility, in order to create a medium of exchange that might reduce the general Zimbabwean preference for pulling everything out of the bank account on payday.
I realise that there is a lot of speculation out there about how much corruption was involved in reaching this point – but I’ll put my head on the block here and say that I suspect there are other more fundamental problems in play:
- Countries that do not have their own currencies are especially vulnerable to fiscal shocks (like droughts and bearish commodity markets and the sudden competitiveness of trading partners due to currency depreciation).
- Countries with weak and almost non-existent domestic productivity are even more vulnerable to fixed exchange rate regimes.
- Countries with a citizenry that does not believe that the banking sector is sound are going to be highly susceptible to cash crises.
- Countries that have to cash in their Balance of Payments current acccounts in order to satisfy a domestic demand for real cash balances are pretty much doomed.
Of course, there are good reasons why Zimbabweans don’t have their own currency, domestic productivity, and faith in the local banking system – so perhaps that’s the bigger issue for some people.
Either way, some observations:
- Zimbabweans seem to be operating under the illusion that the introduction of bond notes will somehow mean that the printing presses will be turned back on. The reality is that the printing presses never stopped creating money in the retail banking sector. This is the definition of fractional banking. The minute a bank extends credit on the back of deposited reserves, they are creating money. And even if they denominate the loans in US dollars – those are still RTGS US dollars that will prove to be unbacked by real/nostro US dollars in a cash crisis. As is now evident.
- A country cannot run sustainably with both a heavy reliance on imports and a cash-based economy. A cash-based economy automatically requires that you only import what you can pay for. As is now also evident.
- Bond notes would simply be an acknowledgement of an already established economic reality, in much the same way that dollarisation in 2009 was an acknowledgement of an already established economic reality.
- And as for this idea that the US dollar cash notes that we’re getting out of the banks now are somehow ‘real’ money, they’re not. Most of the notes in circulation (however limited) are so old and discoloured that they will no longer be accepted by any legitimate bureau de change or bank. They are unusable outside Zimbabwe’s borders – with the possible exception of being exchangeable at a bank in the United States (although even the Federal Reserve has limits on what it will exchange). So unless you’re planning on making a special trip, they’re unusable anywhere but here – and even if you did make that trip, you’re restricted to taking $1,000 at any one time…
And I guess my question is this: if we don’t like the idea of bond notes because we might not be able to use them anywhere other than Zimbabwe, and we’re worried about money printing, and we’re worried that our bank balances won’t be backed by anything real – is that really so different to the existing situation?
All our fears are already fully realised.
And at least the bond notes might be cleaner? But perhaps that’s just me.
PS: seriously, check out the Governor’s statement. It’s illuminating, especially Appendix II.
Rolling Alpha posts opinions on finance, economics, and the corporate life in general. Follow me on Twitter @RollingAlpha, and on Facebook at www.facebook.com/rollingalpha.