Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


Why has the dollar tended to go down as oil goes up?

by Brad Setser
June 18, 2008

I did a podcast for that presents my thinking on this topic.

The simplest reason why oil is up and the dollar is down is that the world economy has been far stronger than the US economy. Weakness in the US economy translates into a weak dollar. Still solid global growth translates into strong demand for oil at time when supplies are a bit tight.

It is also striking, at least to me, that the countries that subsidize oil consumption the most also tend to peg to the dollar or manage their currencies against the dollar. US economic weakness consequently has translated into low US interest rates — and low US rates have translated into low nominal rates – and even lower real rates — in the other, booming dollar zone economies. See Martin Wolf. Combine low real rates with subsidized (or at least below-world-market-prices) oil and there has been a big increase in demand for oil in many countries that peg to the dollar or manage their currencies against the dollar.

I also was persuaded by the analysis of Goldman’s fx team. They argue that there are fundamental reasons to think that a rise in the price of oil should be bad for the dollar. The US economy is energy and oil intensive. The US has the largest existing external deficit of any major oil-importing region. The US exports relatively little to the oil-exporting economies. And the oil-exporting economies seem a bit less inclined to hold dollar-denominated financial assets than in the past.

That said, I wish I had concluded by noting that there are two clear paths that could end the current “oil up, dollar down” pattern.

Weakness in the US economy could drag down global oil demand, pulling both the dollar and oil down. Asia’s 1997-98 crisis led both Asian currencies and the price of oil to fall.

Or a rebound in the US economy could push up the dollar while adding to oil demand. In 2000, a booming US pushed up oil prices and the dollar.

The dollar isn’t always weak when oil is strong. And the dollar isn’t always strong when oil is weak. But so long as global growth is far stronger than US growth, there is reason to think that oil prices will respond to global demand while the dollar will reflect conditions in the US.


  • Posted by crgordon

    It seems logical that as long as oil is priced in dollars that its price will continue to increase with a weaker dollar – that is the only way it will remain unchanged in its purchasing power. Clearly there are other variables impacting price. But I would expect that if all other variables were held constant, the price of oil will move inversely to the dollar.

  • Posted by Dave Chiang

    From Mish Shedlock’s Blog:

    ” Clearly, America, which would largely bear the costs of recession, prefers serial bubble blowing and dollar devaluation as the best policy going forward. The question is whether she can hoodwink, bully, or otherwise convince the rest of the G-8 to go along. The problem is that that the economic stagnation which is evident in the United States is nowhere to be seen in the rest of the world.

    This puts the G-8 delegates into a difficult quandary. If they follow America’s lead to lower interest rates to avoid recession, they risk unleashing inflation, which is already a major problem the world over. If they indicate increased rates, to curb inflation, they risk driving America into a severe recession. Similarly, to engage in a program of currency intervention to support the dollar (which the United States lacks the means or desire to do unilaterally) involves the prospect for even greater accumulation of dollar reserves, which has already proven to be a huge drain on national balance sheets.

    Instead, the G-8 policy statement that emerged at the meeting’s end included no mention of lax lending in the United States or of irresponsible central bank liquidity injections, or even of America’s freeze of on-shore oil drilling. Instead, the G-8 took sideswipes at non-G-8 members, including unbridled criticism of the oil producing countries. U.S. Secretary of State Hank Paulson had the temerity to maintain that higher oil prices were due not just to changes in supply and demand, but also to a failure by oil-rich nations to build enough wells and refineries. ”

  • Posted by satish

    oil is rising because of demand supply mismatch. Oil is not rising due to falling dollar but rising oil price is causing falling dollar. If there is enough spare capacity, why not SA start fresh sell contract in nymex in order of 2000 contracts a day. There will be excess supply and price will crash. Speculators are not going to take delivery. So price at expiry is real price of oil. This is reality, hard to digest but still we have to swallow

  • Posted by ZFC

    Zhou Xiaochuan:

    “A weakening dollar may push up prices of commodities such as crude oil,” which are major imports of China, he said.

  • Posted by glory

    i’d say it’s the natural response to undisciplined fiat currency regimes (viz. negative real rates among ostensible inflation targeters) and on that note, but sorta OT…

    looks like the Fed is extending its tentacles

  • Posted by Michael McKinlay

    Simply Amazing … Not one mention of the $800 Billion Trade deficit.

  • Posted by Rien Huizer

    This one is a bit hard to follow. The logic behind exchange rate movements has yet to be found and investment bank ecomonists are not paid for their contributions to science. To return to your earlier piece on this topic, probably part of the time oil prices influence the USD and part of the time the opposite. I doubt very much that there is a consistent mechanism that makes oil prices and the USD (against what? Euro, Yen, a basket? if so what should be in it?) positively correlated and does so if and only if US growth is (high and) as high as global growth, even if one can show that correlation in the data. Counterfactuals are always a problem in economics, but this one is a little too easy. The current global economic situation is unprecedented in many respects (but so were situations in the past) so there may be millions of plausible explanations for this apparent phenomenon that will remain speculative until we get data that no one wants to share.

    Exchange rates are the supreme insider’s market, with oil a good second..The US has a very long tradition of not intervening in the FX market and quite a few countries shadow the USD for mercantilist purposes. Why the Saudis cs peg is a mystery to me, but at least as long as they do, they take care of a big part of the US trade deficit as a source of weakness. More recently of course (a) the trade deficit has grown on the back of the US/China relationship and (b) Russia end the EU are in the process of becoming a quarrelous partnership on energy. The russians shadow the USD but how long will that be in their interest. So whatever fundamentals may drive your mechanism may also be changing structurally.

    So, lots of interesting and plausible narrarives imaginable, but your theory lacks sufficient foundation.

    That may mean that there may be no problem, or that the problem is there, has no solution but can be pushed away a little longer.or that the sky is about to fall in.

    Incidentally, I found this blog in search of ideas about SWFs but this topic is also interesting. Read today that the Japanese are debating whether or not an SWF should become responsible for investing part of the Gvt pensionfund (the core of the infamous Zaito financial complex). Apparently the Japanese equate an SWF with aggresive, even reckless management..The Reuters report on the discussion is unintelliglible, but one thing in the parliamentary debate stands out: Japan’s reserve management has as one of its constraints not to affront the US. So perhaps there is some form of consultation between the US and the JCB of which the ECB is clearly not capable given its mandate and surrounding politics.

  • Posted by Rien Huizer

    DC, the first two paragraphs of your piece was quite good, why did you add the third one?

  • Posted by Peter

    So, oil up & USD down…. What’s new about it? …… If we follow investment guru like Jim Rogers, he has been saying this for well over 10 years now … BTW, he didn’t have the OPM (other people’s money) syndrome, he had put his own money where his mouth is.

    The only surprise to me is now people are coming out with theories and economic parameters about it.

  • Posted by anon

    “why oil is up”

    In what terms? The nominal dollar price of oil, the real inflation adjusted dollar price of oil, the real exchange rate adjusted dollar price of oil, the Euro price of oil, … ?

    Presumably you mean the nominal dollar price of oil. If so, the outcome isn’t as simple as a choice between a 1:1 or (1:1) relationship.

  • Posted by Dave Chiang

    More of the same Neo-liberalism deregulation of the US Economy for the crony capitalists. Under Goldman Sachs “Bankster” Hank Paulson and Uncle “Bailout” Ben Bernanke, the Enronization of the US Economy continues ….

    Wall Street fights to protect oil speculation

    Wall Street banks and other large financial institutions have begun putting intense pressure on Congress to hold off on legislation that would curtail their highly profitable trading in oil contracts — an activity increasingly blamed by lawmakers for driving up prices to record levels.

    Goldman Sachs lobbyist opposition to the government regulation of Energy trading contracts.

  • Posted by bsetser

    Anon — take your pick: the nominal price of oil in $, the nominal price of oil in euros, the nominal price of oil in SDR (a euro/ $/ pound/ yen basket) or the real price of oil adjusted for US inflation. It doesn’t matter, oil is still up.

    CRGORDAN — the inverse relationship between the dollar and oil over the past two years is true even if oil is denominated in euros or SDR. The effect goes beyond the simple need for the oil to appreciate in dollar terms to keep its euro price constant (or its price v a basket of oil-importing currencies constant).

    Michael — I can hardly be accused of ignoring the trade deficit. Most say I am obsessed by it. One of the reasons I cite for the negative correlation is that the US has the largest existing external (read trade or current account) deficit of any major oil importing region.

    Glory –I put more emphasis on the reasons why high oil (from supply and demand factors) might be bad for the $, but there are also reasons to think that low rates (which contribute to a weak dollar) tend to push up commodity prices. Jeff Frankel has made this argument forcefully.

  • Posted by anon

    Another version of the question is whether oil goes up in dollar terms more or less than the dollar goes down – i.e. whether oil goes up or down in non-dollar terms (e.g. SDR, dollar FX index) – in this case more – or conversely whether the dollar declines more or less than oil increases in dollar terms, in this case less. That nets out the neutral dollar FX effect from the question.

  • Posted by bsetser

    Oil has tended to go up in non-dollar terms (e.g. v SDR or fx index) when the dollar goes down, or perhaps the dollar has tended to go down when oil has gone up v an index. The Goldman fx team’s paper on this (follow the link for a reference to the paper, it isn’t publicly available) is quite good.

  • Posted by Dave Chiang

    The real US Economy is collapsing. Auto sales are dropping to Depression-like levels. Bailouts for Wall Street, but not for the real US Economy:

    “From the Detroit Free Press:

    * After months of confident talk that Chrysler LLC anticipated the economic downturn better than other automakers, Chairman and Chief Executive Officer Bob Nardelli told employees Tuesday, in a memo obtained by the Free Press, that the last few months — and this month in particular — have been even worse than Chrysler anticipated.

    * So far in June, he said, J.D. Power and Associates and Citigroup are seeing a sales pace that is almost 20% lower — only 12.5 million vehicles per year.

    * “We thought we were being extremely aggressive in our conservative view” of 2008, Nardelli said late last month. “As it turns out we may have been spot-on.”

    * If J.D. Power’s forecast for June — an annualized rate of 12.5 million sales — continues for long, Erich Merkle of IRN Inc. said, it would be “Armageddon. Doomsday.”

    * “I don’t think there is anyone out there prepared for 12.5″ million annual U.S. sales, he said. “I know it is only one month, but still that would show some signs that there is some real deterioration in the market, that would mean the economy is really slowing down significantly.””

    If this trend continues, the collapse of the Big 3 is inevitable, as they cannot re-tool fast enough to produce enough (low profit)sub-compact gas-sippers to make up for lost big pickup/SUV profits. Chrysler, GMC and Ford will fail in that order, although it is a real tight race to the bottom. Toyota and Honda are laughing, while the rest weep.

  • Posted by Dave Chiang


    Iraqi Energy Ministry awards no-bid energy development contracts to US Corporations. Chinese and Russian oil contracts voided.

    Deals With Iraq Are Set to Bring U.S. Oil Giants Back

    BAGHDAD — Four Western oil companies are in the final stages of negotiations this month on contracts that will return them to Iraq, 36 years after losing their oil concession to nationalization as Saddam Hussein rose to power.

    The deals, expected to be announced on June 30, will lay the foundation for the first commercial work for the major companies in Iraq since the American invasion, and open a new and potentially lucrative country for their operations.

    The no-bid contracts are unusual for the industry, and the offers prevailed over others by more than 40 companies, including companies in Russia, China and India.

  • Posted by Realist

    Oil is up and the USD down because of separate (though not mutually independent) reasons, which in fact are the same for each: the respective supply and demand fundamentals.

    That the oil price rise – which has also happened in terms of stronger currencies such as EUR and CAD, and even in terms of gold – is due to supply/demand balance can be seen from the IEA monthly reports. Moreover, the latest price rise could already be forecasted out of the demand projections for 2008 in the June 2007 report (which were based on the oil price back then; current demand projections are lower because higher prices have caused demand destruction.)

    And the weakness of the USD has no DIRECT connection to weakness in the US economy. The USD is weak just because there are too many dollars around. The connection is indirect: weakness in the US economy prompts the Fed to loosen US monetary policy and increase the supply of dollars (via REPOs, TAFs, etc.) which is what in turn weakens the USD. But if (hypothetically) the Fed stood tight, there could have been a deep recession in the US AND a strong dollar (path a. below).

    The interdependency of the oil and USD supply/demand fundamentals comes from the fact that the US is the biggest oil importer. Therefore an oil price rise tends to increase the already large US trade and current account deficits. More dollars abroad push the USD value down.

    And the two mentioned paths that could end the current “oil up, dollar down” pattern are just not realistic. The realistic paths are:

    a. Weakness – indeed, a deep recession – in the US economy coming as a result of a substantial tightening of US monetary policy would lead to an “oil down, dollar up” outcome. Which does not mean that the whole world would enter a recession and the oil price would plunge. Both China/East Asian countries and the oil exporters have accumulated such huge forex reserves over the last years that their situation is entirely different from that in 2001: having already saved for a rainy day (or rather decade), they can now afford to keep growing their internal consumption even if OECD countries enter recession.

    b. A rebound in the US economy as a result of monetary and fiscal stimulus will just lead to a continuation of the current “oil up, dollar down” pattern, since it will just sustain US oil demand and thus increase the amount of dollars abroad. As for the fact that “in 2000, a booming US pushed up oil prices and the dollar”, in 2000 the US trade and CA deficit were much smaller, the US Federal Gov had a huge fiscal surplus and was on the way to repaying all federal debt, US Fed Funds rates were higher, China/East Asian countries and the oil exporters had nowhere near the current levels of forex reserves, the oil market was in a completely different state, and there was still the illusion that high tech (particularly Internet-related) was the bright future and commodities were the dull past. Quite a different scenario from now.

    c. “The third horseman path”: the US accelerates the diversion of corn to ethanol, and soybeans to biodiesel, causing a sharp reduction of US agricultural exports in volume and their sharp increase in dollars (through skyrocketing agricultural prices), thus balancing the US CA deficit. The resulting “oil up, dollar up, grains to the moon” outcome would be beyond Malthus’ worst nightmare.

  • Posted by Dave Chiang

    Pending Israeli attack on Iran will see Middle East in Flames – Oil over $200 per barrel
    ‘MISSION DOABLE’,1518,559925,00.html

    The Israeli government no longer believes that sanctions can prevent Iran from building nuclear weapons. A broad consensus in favor of a military strike against Tehran’s nuclear facilities — without the Americans, if necessary — is beginning to take shape.

  • Posted by Dave Chiang

    China hikes prices a tiny bit and the price of Oil gets hammered big time … yeah … that right … Goldman Sachs Paulson says it’s supply and demand fundamentals … LOL

  • Posted by Dave Chiang

    Federal Reserve holding $50 billion of Iraqi government’s money as ransom to control Iraq’s oil wealth

    However, on Thursday, the New York Times reported that Exxon Mobil, Shell, Total and BP were in the final stages of negotiations on contracts that will return them to Iraq, 36 years after losing their oil concession to nationalization by Saddam Hussein.

    They are reportedly in negotiations with the Oil Ministry for no-bid contracts to service Iraq’s largest fields. Should the deals go through, they would lay the foundation for the first commercial work for major Western companies in Iraq since the American invasion in 2003. It is expected that Iraq’s output could increase to about 3 million barrels a day from its current 2.5 million.

    Initially, the Bush administration wanted no less than 58 permanent US bases in Iraq. There are already 30 in place. It doesn’t matter that on April 8, US ambassador to Iraq Ryan Crocker had said the US “will not establish permanent bases in Iraq and we anticipate that it will expressly foreswear them”.

    The Bush administration’s ploy essentially amounts to turning over legal control of US bases to a client regime. Heavy pressure is the name of the game. To convince the Iraqis, the Bush administration is holding no less than $50 billion of Iraqi money in the Federal Reserve Bank of New York. Other “subtle” forms of pressure also apply. The Iraqis wanted to sell oil in euros as well as in dollars. The Bush administration issued its fatwa – and it’s a “no”.

  • Posted by Thomas

    Oil is the new gold. Dollar, gold, oil, land: which would you rather hold for the next 5 years?

  • Posted by bsetser

    Realist — the US fed data (factors affecting reserve balances — basically the fed’s balance sheet) doesn’t suggest that the fed has been printing a lot of dollars. indeed, the absence of more expansion in the absolute size of the fed’s balance sheet during this rate cutting cycle has been a bit surprising. the number of dollars circulating hasn’t increased.

    what the fed has done is released of lot of its stock of treasuries into the market by lending out treasuries to banks and broker dealers against less liquid collateral (mostly agency bonds i think). so there is a much stronger argument that the us is flooding the world with treasury bonds than that it is flooding the world with dollars.

    central banks outside the us by contrast have increased the size of their balance sheet quite significantly — and often have experienced faster growth in base money.

    I’ll grant that US rates are low. but I don’t see a rapid increase in base money or the overall size of the fed’s balance sheet.

  • Posted by crgordon

    Please ignore this question if the answer is too silly obvious. I read that the Fed has not increased the money supply by virtue of swapping Treasuries for less liquid agencies. So net, there is no increase of dollars in circulation. I get that. However, if there is a huge transfer of wealth to the Gulf countries for oil and to China for manufactured product, from whence do these dollars spring? There isn’t enough currency in circulation so payment is not in physical currency. But it is a transaction in “dollars”. Is the source of these “dollars” emanating from the eternal spring of issuing more Treasuries? And, if so, isn’t this an expansion of the money supply by the USG instead of the Fed, the operative word being “expansion”? Into what do these Treasuries eventually morph? And whatever those are, are they not things that are dollar equivalents? Aren’t Treasuries the USG equivalent of printing money?

  • Posted by Rien Huizer

    I like Realist’s argument intuitively but agree that the Fed data do not support it (yet). However it is one of many a prima facie plausible explnations.

    In general, it is more likely that any correlations between the USD/EEEO (everything else except oil) and the Oil/EEEO are based on a vast and opaque complex of factors that vary in strength and direction some of which due to policy, some randomly and autonomously. With positive and negative feedback thoughout the system. Some market participants (centrl banks, governments, oil producers and majors) may have a short term information advantage (they know their own next moves and have the right size). In a case like that it is safe not to assume that historic patterns of correlation will hold, or that recent variations introduce a new trend. Pity for journalists and investment banking analysts. When I was young we studied macroeconomics and were made to believe that we could design the economy because we had computer models. We ended to forget that the world is full of humans, politicians and businessmen..

  • Posted by anon

    You won’t find the answer to any of this by examining the supply of dollars.

    Monetarism is dead.

  • Posted by rebeleconomist


    Perhaps I can answer your question. The dollars that the US paid for its imports have been recovered by selling financial assets to foreigners, so that the net result has been (largely) an exchange of debt for goods. The fact that the debt is denominated in dollars means that the US has promised to pay a specified number of dollars to foreigners, who may then use them to make either renewed purchases of US assets or US goods and services. But those dollars have not been printed yet. And if the debtor raises the dollars by selling something, or in the case of the government, taxing something (effectively selling a government service), it will not be necessary for the US to print more dollars in the future. Of course, the temptation for the US will be to print more dollars and cheat the foreign creditors by giving them something worth less than expected.

  • Posted by crgordon

    Thank you for the explanation and the concept of taxes being a sale of government service. A note about the quality of financial assets being sold: The structure of some seem to be the modern day alchemist’s answer of turning lead (debt) into gold. Caveat Emptor.

  • Posted by flow5

    “the US fed data (factors affecting reserve balances — basically the fed’s balance sheet) doesn’t suggest that the fed has been printing a lot of dollars. indeed, the absence of more expansion in the absolute size of the fed’s balance sheet during this rate cutting cycle has been a bit surprising. the number of dollars circulating hasn’t increased.”

    Legal reserves are no longer binding/constraining.

    Inflation is always and everywhere a monetary phenomenon. Friedman was right in this respect.+

  • Posted by flow5

    I.e., the FED is following an EXTREMELY easy monetary policy.

  • Posted by anon

    “But those dollars have not been printed yet.”

    So the velocity of money is 0?

  • Posted by flow5

    There are very few (if any) people who understand money & central banking. The growth of “loans-deposits” have ranged from a +13% rate-of-change to a +9% roc. I.e., “money supply” growth has been on an extremely inflationary path.

    Neither Friedman’s “high powered money” [sic], nor legal reserves have a predictable “expansion coefficient”. “I. e., there is general agreement that, for almost all banks throughout the world, statutory reserve requirements are not binding, Richard G. Anderson, V. P. & Economist – the Maverick Bank”.

    Since the “money multiplier” has been emasculated, what the net expansion of money will be as a consequence of a given injection of additional reserves won’t be known until long after the fact.

    Contrary to the Keynesian economists that dominate the Fed’s research staff, the money supply can never be managed by any attempt to control the cost of credit.

  • Posted by Claire

    Nice article and comments. That was what I was looking for. Thanks to all of you.