One claim that has been made time and again in Ethiopian economic discourse over the past five years or so is that “printing money” on the National Bank of Ethiopia’s (NBE) part has caused high inflation. And people making this assertion seem to have been vindicated by recent events: the federal government has refrained from taking direct advances from the NBE for two consecutive years and, meanwhile, official numbers have begun to show easing in the inflation rate (up until the war in Iran). But is that claim defensible on factual ground? Is it even reflective of systematic economic thinking? I think the answer to both questions is no.
Making spurious and overblown assertions about inflation in Ethiopia is certainly not new. During the Ethiopian People’s Revolutionary Democratic Front (EPRDF) era, when the economy was officially growing by double digits, some people, including public officials, contended that observed inflation was the result of an “overheated economy,” either failing to appreciate how this would make little sense in an economy characterized by considerable slack or just trying to dupe the public. There have also been times when we heard the bizarre argument that inflation caused by “man-made” factors is unacceptable, as if price increases ought to be a natural phenomenon. So the alleged inflation driver at hand, i.e. printing money, is only one among many delusions.
Why should we care? Mainly, it’s because inappropriate diagnosis of an economic problem, especially one made by elite opinion-makers or institutions which hold sway over government policy, can lead to inappropriate treatment, which not only leaves root causes unaddressed, but may even be counterproductive.
Now what do we mean by printing money? Strictly speaking, it refers to the actual issuance of new currency by a government. In practice, however, direct central bank financing of government deficits, i.e. monetizing the debt, is also commonly, if less accurately, referred to as printing money. But never mind. Whichever notion people have in mind when they talk about printing money, the theoretical effect of the two cases is essentially the same, which is to increase the monetary base (i.e. create high-powered money) and, thereby, the money supply.
So let’s be clear: Printing money does not necessarily cause inflation. Yes, there is a general economic principle stating that high inflation, especially hyperinflation, arises when the government prints too much money, or when there is excessive money supply growth. But there are so many ifs and buts here. For example, how much is “too much”? How large and for how long is government spending covered by printing money? And how close is the economy running to its productive capacity? To propose a monetary explanation for inflation even as a hypothesis requires answering these questions. By contrast, there is a pervasive inclination among Ethiopian economic observers to assume any monetary injection as inflationary, as if the economy is in equilibrium at full employment – which it isn’t, not even remotely.
More important, the money-printing-turned-inflation story does not fit the available data at all. First of all, it’s not like the government has financed deficits in or shocks to its budget solely with money creation; at least part of government spending has been covered by domestic and external debt. Also, neither the money supply nor inflation has risen in an extreme fashion. By the NBE’s own reckoning, the M2 money supply increased at an average rate of 26% a year between 2009/10 and 2023/24 – hardly extreme growth! And, for what it’s worth, Ethiopia’s M2/GDP ratio remains among the lowest in the world. Inflation, too, averaged 17% per annum, which is clearly excess, and hurtful, but by no means hyperinflation. Then, if you plotted the money supply (its growth rate) and the consumer price level (its inflation rate) over the same 15-year period, you would not observe tightly linked co-movements. And this should be no surprise given our dysfunctional interest rate and credit channels of monetary transmission.
But here is an even more informative analysis: The NBE’s control is confined to the monetary base – currency plus total reserves in the banking system – or to what we call reserve money – currency plus bank deposits at the central bank. And there were striking discrepancies in the growth trajectories of reserve money and the M2 money supply over 2009/10-2023/24. On some occasions, when the growth of reserve money plunged, money supply recorded double-digit growth – the two giving conflicting signals about the course of monetary policy. On others, when the expansion of reserve money became more rapid, that of money supply got slower. How, then, can you blame the NBE’s money-printing for inflation when even reserve money’s causal influence on the money supply has been dubious? Viewing the NBE as a sole source of money supply growth – which is also affected by what banks, depositors and borrowers do – or telling it to simply reduce the amount of money in the economy is indeed foolish. And in part, at least, the money growth probably reflects the fact that, as our economy has expanded, so has the money people need to hold for business transactions.
Note also that I haven’t even raised the issue of whether or not the NBE could actually justify an expansionary monetary policy stance – which it could in an economy suffering from high unemployment of both workers and resources.
At this point you might be wondering why the NBE’s revised establishment proclamation has then tightened the constraint on direct monetary financing of budget deficits (no more than 15% of average annual government revenues of the previous three years and only in the form of temporary overdrafts), notwithstanding the workarounds the federal government could still find. The compelling reason is that it helps to limit fiscal dominance of monetary policy and discourage fiscal irresponsibility. Similarly, the two-year-old termination of direct advances to the government is better understood as a precautionary measure than as a remedial action warranted by hard evidence of past inflation driven by excessive money supply.
But then, what would a more plausible account of our inflation look like? Surely it includes multiple factors, both from the demand and the supply sides, and not all domestic. Even the IMF – the ultimate producer of monetary narratives of inflation – identified drought-induced agricultural output shocks, currency devaluation, administered-price increases, political disorder that disrupted supply chains, and world food and commodity price surges as factors explaining Ethiopia’s excess inflation between 2009/10 and 2023/24. We also had a two-year-long war initiated in Tigray, as well as imported inflation due to the Covid-19 pandemic and especially the Russia-Ukraine war. Add to this the high likelihood that a rise in inflation expectations has played a role in pushing up actual prices lately, and characterization of our inflation as an offspring of money-printing ends up being a figment of the imagination.
One has to wonder, though, why some economic commentators have made a causal conjecture contrary to sound analytical reasoning and available evidence. It is safe to say that most of them have not thought the matter through. From a general knowledge of the government’s money-printing to pay some of its bills, they have made a very large leap to blame that for visible inflation, overlooking the economic circumstances under which this may be true or not bothering to check if their position holds in the data. Beyond this, some economic pundits probably cannot resist the urge to tell a monetary story about inflation because doing so sounds advanced. Whatever the real motivation, some soul searching is in order.
And not just among the commentator group. Just last month, an academic paper presented at the 23rd International Conference on the Ethiopian Economy singled out, once again, monetary expansion for driving domestic inflation. And based on this spurious inference, the study made audacious medium-term projections for inflation, before recommending that the NBE cap the growth of money supply in order to achieve single-digit inflation. Hah. If only matters were that simple and our inflation was that predictable.
That’s not the end of the story, though, since the claim that money-printing has caused inflation sometimes gets transmuted into declaration that tight monetary policy has enabled inflation to fall. The latter assertion all too often suffers from idle speculation, in three respects. First, it apparently commits the false cause fallacy: thinking that, because disinflation comes together with or right after the NBE’s monetary tightening, there must be a causal connection, without properly establishing the latter. Second, it does not tell us under which empirically founded transmission mechanism tight money causes disinflation in Ethiopia – presumably because there isn’t one. Finally, this assertion betrays the misguided belief that tight money directly brings inflation down, just like a tree cutter, whereas it first requires weakening economic activity or growth, and thereby possibly employment – the evidence of which is usually lacking.
But doesn’t the NBE’s Monetary Policy Committee (MPC) meet regularly and make proposals to manipulate liquidity in the banking system, especially commercial banks’ credit supply, with a view to achieving price stability? Well, policy must be made and the committee must rationalize whatever stance it adopts. My point is that at least in the foreseeable future its decisions can at most affect particular sectors of the economy, falling short of substantively influencing inflation due to lack of credible transmission channels on the ground. Thus it is logical for bank-dependent borrowers, be it individuals or businesses, to wonder if the MPC’s decisions impact their desired cash and working capital. But well-informed, policy-oriented macroeconomists are unlikely to hold their breath.
In short, thinking of our inflation as a monetary phenomenon is far too simplistic. The claims purporting to have found a strong link between money printing/supply and inflation do not stand up to closer scrutiny. And you really don’t want to judge these claims by their sources, i.e. by how elite the latter are or appear to be. Only thinking things through and paying enough attention to factual conformity reveal that monetary factors have been of minor importance to aggregate economic activity and inflation in Ethiopia. And that is the bottom line.
