The main conclusion remains the same though: unless Japan’s fiscal policy changes, or the US seriously cuts interest rates, and/or Japan’s economic fundamentals change, an intervention will probably not be very effective.
In a simple model where interest rate gap is the main factor behind exchange rate moves:
(Referring back to my original metaphor of water) Water will flow as long as there is a significant difference in level (not only when the difference increases).
Therefore I would think that the area in the rectangular on your chart is as expected (and not a yen weaker than expected/predicted).
And indeed we saw the same in the period 2012-2015, when gap was high but did not increase (actually slightly decreased) and yet the yen kept going down.
As far as I can see the the top chart only covers just over 2.5 years and not 12 years.
The difference in how we view things seems to be in what happens when the interest rate gap is persistently high (but doesn’t grow further).
You describe the area inside the triangle of your top chart as “weaker than one would predict”.
In my view, in a simple model where the interest rate gap is the main factor driving exchange rate, the exchange rate would continue to deteriorate when the interest rate gap stays high but doesn’t increase, and accordingly the for the area within the oval circle of your chart, the yen behaves as should be expected based on a simple model where interest rate gap is the main factor driving the USDJPY exchange rate. (In answer to your question: the ‘force’ being “money is lured from Japan to the US. Those purchases of US financial assets create demand for dollars” as described by you.)
The same was also the case for roughly the period 2012–2015, where the interest rate gap was wide but didn’t increase (actually slightly decreased), and the yen deteriorated from 77 to 120.
If I understand you correctly, your argument is that whenever a country has a capital outflow due to any interest rate gap, that will cause its currency to depreciate, Not necessarily. It is a staple of macroeconomics that countries with current account surpluses (trade and net income from abroad on past investments) will have a capital outflow of an equal amount. And those with current account deficits will have a capital inflow. In essence, those with deficits are borrowing from the rest of the world to finance the deficit and those with surpluses are financing others' purchases of their net exports. The question is: at what price do these two come into balance? If ex ante, the capital outflow would be too large, then the currency will depreciate to raise the current account. If ex ante they are in balance, even with a gap in interest rates, then the currency rate need not change.
“For medium-term periods, there is often a very close correlation between MOVEMENTS in the interest rate gap and MOVEMENTS in the level of the yen/$.”
I am not surely that necessarily holds true. If going forward the interest rate gap would stay steady at say 3.5% or 5% (and doesn’t move from there), wouldn’t you expect the yen to continue to depreciate? (Even in medium term.)
What force(s) would cause it to depreciate indefinitely, with some interruptions? To my mind, there are so many conflicting forces in play? After all, in the bottom chart, the yen moved in line with the rate gap for 12 years, a long time, in such matters with an 80% R-squared. So, let me know what forces you see causing this depreciation and whether you see this going on for years, or decades or what?
First of all, I really enjoyed your article. You made a lot of good points, and I think your conclusion is valid.
I only did not fully understand your first graph. If I understand it correctly, you are plotting a timeline for exchange rates against the difference between interest rates.
I think a (flawed but still useful) metaphor for a very simplified model for the relationship between interest rates and exchange rates is two interconnected vessels of water, with water flowing from one vessel to the other as long as there is a difference in water level between the two vessels (the usual metaphor is water flowing from a higher point to a lower point, so you can use that also, but I find a metaphor of vessels working better). Water would flow from vessel A to vessel B as long as the water level in vessel A is higher than the water in vessel B = if there is a positive difference in water level between vessel A and vessel B. Water would flow the other way if the that difference becomes negative. And the speed of water flowing between the two vessels would show a positive correlation with the differential of water level.
In this metaphor the two vessels are two countries, water flow is capital flow between the countries, and the difference in water level is difference in interest rates.
You would expect exchange rates to go up as long as there is a positive difference between the interest rates of the two countries, and exchange rates go up faster when the difference becomes bigger, a relationship which your graphs of the two timelines mostly (though not fully, it is a very simplified model and many other factors play a role) conforms to. However, if the interest rate gap no longer increases, but would stay constant, for example at a delta of 3.5%, you would expect the two lines to diverge: while the grey line would stay at the same level, the black would keep going up (as you indeed start seeing at the end of your time series). Similarly, if the grey line would go down, the black line would be generally be expected to keep going up (though probably at a lower speed), until the black line gets (close) to the zero level.
Has become a bit a long story, but is the fact that those two timeseries move so nicely together not mostly an artifact of (a) it being a short time series during which (b) the interest rate gap always being positive during the timeseries (c) the interest rate gap mostly moving in one direction during the timeseries?
Again, this is just trying to understand what relationship you are trying to express with the first graph (and how you subsequentially derive an exchange rate forecast from it*) and not intended to take anything away from all the good points you make or the validity of your conclusion.
And my understanding might very well be wrong, or I might have misunderstood what you were trying to express. Would be interested in your thoughts.
(*While interest rate gap the can more or less predict the direction of exchange rates (or at least is major factor), I don’t think absolute levels of future exchange rates can be derived from or predicted from interest rate differentials.)
Peter, thanks for your comment. I think the answer to your question lies in the contrast between the chart at the top and the one at the bottom. For medium-term periods, there is often a very close correlation between MOVEMENTS in the interest rate gap and MOVEMENTS in the level of the yen/$. But, as we see over time--e.g. the difference in the absolute level between 2001-13 and 2021-24 in the bottom chart--the absolute level can change due to factors more fundamental than the interest rate gap. This is something I've referred to in post after post. That absolute level is indicated by the constant in the equation in the bottom chart. If there were no gap at all in interest rates, the yen would have been Y65 in 2001-2013 and Y88 in 2021-24. All along the two trend lines, the yen is about 22 points weaker these days than in 2001-13. So, in 2024 we see a weaker yen at any given interest rate gap than we saw in 2001-13. As i discussed, that longer change is due to the deterioration of fundamentals, like relative productivity and competitiveness and relative price levels, over the long haul. I hope this answers your question.
I am a farmer. I like Iseki tractors. The price in Australia is fixed by local agents. Perhaps the Japanese need to be be proactive in overseas markets in ensuring that their great products are more competitively priced.
the 2001-13 period is a straw man argument because of all the others factors in play especially following the GFC in which the world was in a risk off environment.The Japanese financial system has massive amounts of money invested around the globe which results in repatriation when the global system is under stress.Also,the continued flawed policy of the MOF/BOJ keeps the Yen in play as the premier vehicle of the "carry trade" which weakens the YEN beyond its domestic fundamentals.The Chinese are not happy seeing the Yen/Yuan cross at levels above December,1993 and June,1998 ----in 1998 the US TREASURY intervened to BUY YEN at the behest of the Chinese when Bill Clinton was going to visit Beijing .So what is the global financial policy makers intending to do especially as the eur/yen is far too strong for the German economy to compete with the Japanese----the eur/yen is also back to that important level of June,1998
The main conclusion remains the same though: unless Japan’s fiscal policy changes, or the US seriously cuts interest rates, and/or Japan’s economic fundamentals change, an intervention will probably not be very effective.
In more simple words:
In a simple model where interest rate gap is the main factor behind exchange rate moves:
(Referring back to my original metaphor of water) Water will flow as long as there is a significant difference in level (not only when the difference increases).
Therefore I would think that the area in the rectangular on your chart is as expected (and not a yen weaker than expected/predicted).
And indeed we saw the same in the period 2012-2015, when gap was high but did not increase (actually slightly decreased) and yet the yen kept going down.
As far as I can see the the top chart only covers just over 2.5 years and not 12 years.
The difference in how we view things seems to be in what happens when the interest rate gap is persistently high (but doesn’t grow further).
You describe the area inside the triangle of your top chart as “weaker than one would predict”.
In my view, in a simple model where the interest rate gap is the main factor driving exchange rate, the exchange rate would continue to deteriorate when the interest rate gap stays high but doesn’t increase, and accordingly the for the area within the oval circle of your chart, the yen behaves as should be expected based on a simple model where interest rate gap is the main factor driving the USDJPY exchange rate. (In answer to your question: the ‘force’ being “money is lured from Japan to the US. Those purchases of US financial assets create demand for dollars” as described by you.)
The same was also the case for roughly the period 2012–2015, where the interest rate gap was wide but didn’t increase (actually slightly decreased), and the yen deteriorated from 77 to 120.
I meant the bottom chart, during 2001-13.
If I understand you correctly, your argument is that whenever a country has a capital outflow due to any interest rate gap, that will cause its currency to depreciate, Not necessarily. It is a staple of macroeconomics that countries with current account surpluses (trade and net income from abroad on past investments) will have a capital outflow of an equal amount. And those with current account deficits will have a capital inflow. In essence, those with deficits are borrowing from the rest of the world to finance the deficit and those with surpluses are financing others' purchases of their net exports. The question is: at what price do these two come into balance? If ex ante, the capital outflow would be too large, then the currency will depreciate to raise the current account. If ex ante they are in balance, even with a gap in interest rates, then the currency rate need not change.
“For medium-term periods, there is often a very close correlation between MOVEMENTS in the interest rate gap and MOVEMENTS in the level of the yen/$.”
I am not surely that necessarily holds true. If going forward the interest rate gap would stay steady at say 3.5% or 5% (and doesn’t move from there), wouldn’t you expect the yen to continue to depreciate? (Even in medium term.)
What force(s) would cause it to depreciate indefinitely, with some interruptions? To my mind, there are so many conflicting forces in play? After all, in the bottom chart, the yen moved in line with the rate gap for 12 years, a long time, in such matters with an 80% R-squared. So, let me know what forces you see causing this depreciation and whether you see this going on for years, or decades or what?
First of all, I really enjoyed your article. You made a lot of good points, and I think your conclusion is valid.
I only did not fully understand your first graph. If I understand it correctly, you are plotting a timeline for exchange rates against the difference between interest rates.
I think a (flawed but still useful) metaphor for a very simplified model for the relationship between interest rates and exchange rates is two interconnected vessels of water, with water flowing from one vessel to the other as long as there is a difference in water level between the two vessels (the usual metaphor is water flowing from a higher point to a lower point, so you can use that also, but I find a metaphor of vessels working better). Water would flow from vessel A to vessel B as long as the water level in vessel A is higher than the water in vessel B = if there is a positive difference in water level between vessel A and vessel B. Water would flow the other way if the that difference becomes negative. And the speed of water flowing between the two vessels would show a positive correlation with the differential of water level.
In this metaphor the two vessels are two countries, water flow is capital flow between the countries, and the difference in water level is difference in interest rates.
You would expect exchange rates to go up as long as there is a positive difference between the interest rates of the two countries, and exchange rates go up faster when the difference becomes bigger, a relationship which your graphs of the two timelines mostly (though not fully, it is a very simplified model and many other factors play a role) conforms to. However, if the interest rate gap no longer increases, but would stay constant, for example at a delta of 3.5%, you would expect the two lines to diverge: while the grey line would stay at the same level, the black would keep going up (as you indeed start seeing at the end of your time series). Similarly, if the grey line would go down, the black line would be generally be expected to keep going up (though probably at a lower speed), until the black line gets (close) to the zero level.
Has become a bit a long story, but is the fact that those two timeseries move so nicely together not mostly an artifact of (a) it being a short time series during which (b) the interest rate gap always being positive during the timeseries (c) the interest rate gap mostly moving in one direction during the timeseries?
Again, this is just trying to understand what relationship you are trying to express with the first graph (and how you subsequentially derive an exchange rate forecast from it*) and not intended to take anything away from all the good points you make or the validity of your conclusion.
And my understanding might very well be wrong, or I might have misunderstood what you were trying to express. Would be interested in your thoughts.
(*While interest rate gap the can more or less predict the direction of exchange rates (or at least is major factor), I don’t think absolute levels of future exchange rates can be derived from or predicted from interest rate differentials.)
Peter, thanks for your comment. I think the answer to your question lies in the contrast between the chart at the top and the one at the bottom. For medium-term periods, there is often a very close correlation between MOVEMENTS in the interest rate gap and MOVEMENTS in the level of the yen/$. But, as we see over time--e.g. the difference in the absolute level between 2001-13 and 2021-24 in the bottom chart--the absolute level can change due to factors more fundamental than the interest rate gap. This is something I've referred to in post after post. That absolute level is indicated by the constant in the equation in the bottom chart. If there were no gap at all in interest rates, the yen would have been Y65 in 2001-2013 and Y88 in 2021-24. All along the two trend lines, the yen is about 22 points weaker these days than in 2001-13. So, in 2024 we see a weaker yen at any given interest rate gap than we saw in 2001-13. As i discussed, that longer change is due to the deterioration of fundamentals, like relative productivity and competitiveness and relative price levels, over the long haul. I hope this answers your question.
I am a farmer. I like Iseki tractors. The price in Australia is fixed by local agents. Perhaps the Japanese need to be be proactive in overseas markets in ensuring that their great products are more competitively priced.
the 2001-13 period is a straw man argument because of all the others factors in play especially following the GFC in which the world was in a risk off environment.The Japanese financial system has massive amounts of money invested around the globe which results in repatriation when the global system is under stress.Also,the continued flawed policy of the MOF/BOJ keeps the Yen in play as the premier vehicle of the "carry trade" which weakens the YEN beyond its domestic fundamentals.The Chinese are not happy seeing the Yen/Yuan cross at levels above December,1993 and June,1998 ----in 1998 the US TREASURY intervened to BUY YEN at the behest of the Chinese when Bill Clinton was going to visit Beijing .So what is the global financial policy makers intending to do especially as the eur/yen is far too strong for the German economy to compete with the Japanese----the eur/yen is also back to that important level of June,1998