宏观经济运行的框架（英文原版）by_Ray_Dalio--Bridgewater

© 2012 Bridgewater Associates, LP 1 A Template for Understanding… …How the Economic Machine Works and How it is Reflected Now Ray Dalio Created October 31, 2008 | Updated March, 2012 The economy is like a machine. At the most fundamental level it is a relatively simple machine, yet it is not well understood. I wrote this paper to describe how I believe it works. My description is not the same as conventional economists’ descriptions so you should decide for yourself whether or not what I’m saying makes sense. I will start with the simple things and build up, so please bear with me. I believe that you will be able to understand and assess my description if we patiently go through it. How the Economic Machine Works: “A Transactions-Based Approach” An economy is simply the sum of the transactions made and a transaction is a simple thing. A transaction consists of the buyer giving money (or credit) to a seller and the seller giving a good, a service or a financial asset to the buyer in exchange. A market consists of all the buyers and sellers making exchanges for the same things – e.g., the wheat market consists of different people making different transactions for different reasons over time. An economy consists of all of the transactions in all of its markets. So, while seemingly complex, an economy is really just a zillion simple things working together, which makes it look more complex than it really is. For any market, or for any economy, if you know the total amount of money (or credit) spent and the total quantity sold, you know everything you need to know to understand it. For example, since the price of any good, service or financial asset equals the total amount spent by buyers (total $) divided by the total quantity sold by sellers (total Q), in order to understand or forecast the price of anything you just need to forecast total $ and total Q. While in any market there are lots of buyers and sellers, and these buyers and sellers have different motivations, the motivations of the most important buyers are usually pretty understandable and adding them up to understand the economy isn’t all that tough if one builds from the transactions up. What I am saying is conveyed in the simple diagram below. This perspective of supply and demand is different from the traditional perspective in which both supply and demand are measured in quantity and the price relationship between them is described in terms of elasticity. This difference has important implications for understanding markets. Reason A Reason B Reason C Buyer 1 Reason A Reason B Reason C Buyer 2 Reason A Reason B Reason C Buyer 3 Reason A Reason B Reason C etc… Reason A Reason B Reason C Seller 1 Reason A Reason B Reason C Seller 2 Reason A Reason B Reason C Seller 3 Reason A Reason B Reason C etc… Total Q Total $ Price = Total $ / Total Q Total $ = Money + Credit Reason A Reason B Reason C Reason A Reason B Reason C Buyer 1 Reason A Reason B Reason C Reason A Reason B Reason C Buyer 2 Reason A Reason B Reason C Reason A Reason B Reason C Buyer 3 Reason A Reason B Reason C Reason A Reason B Reason C etc… Reason A Reason B Reason C Reason A Reason B Reason C Seller 1 Reason A Reason B Reason C Reason A Reason B Reason C Seller 2 Reason A Reason B Reason C Reason A Reason B Reason C Seller 3 Reason A Reason B Reason C Reason A Reason B Reason C etc… Total Q Total $ Price = Total $ / Total Q Total $ = Money + Credit 2 The only other important thing to know about this part of the Template is that spending ($) can come in either of two forms – money and credit. For example, when you go to a store to buy something you can pay with either a credit card or cash. If you pay with a credit card you have created credit, which is a promise to deliver money at a later date,1 whereas, if you pay with money, you have no such liability. In brief, there are different types of markets, different types of buyers and sellers and different ways of paying that make up the economy. For simplicity, we will put them in groups and summarize how the machine works. Most basically:  All changes in economic activity and all changes in financial markets’ prices are due to changes in the amounts of 1) money or 2) credit that are spent on them (total $), and the amounts of these items sold (total Q). Changes in the amount of buying (total $) typically have a much bigger impact on changes in economic activity and prices than do changes in the total amount of selling (total Q). That is because there is nothing that’s easier to change than the supply of money and credit (total $).  For simplicity, let’s cluster the buyers in a few big categories. Buying can come from either 1) the private sector or 2) the government sector. The private sector consists of “households” and businesses that can be either domestic or foreign. The government sector most importantly consists of a) the Federal Government2 which spends its money on goods and services and b) the central bank, which is the only entity that can create money and, by and large, mostly spends its money on financial assets. Because money and credit, and through them demand, are easier to create (or stop creating) than the production of goods and services and investment assets, we have economic and price cycles. The Capitalist System As mentioned, the previously outlined economic players buy and sell both 1) goods and services and 2) financial assets, and they can pay for them with either 1) money or 2) credit. In a capitalist system, this exchange takes place through free choice – i.e., there are “free markets” in which buyers and sellers of goods, services and financial assets make their transactions in pursuit of their own interests. The production and purchases of financial assets (i.e., lending and investing) is called “capital formation”. It occurs because both the buyer and seller of these financial assets believe that the transaction is good for them. Those with money and credit provide it to recipients in exchange for the recipients’ “promises” to pay them more. So, for this process to work well, there must be large numbers of capable providers of capital (i.e., investors/lenders) who choose to give money and credit to large numbers of capable recipients of capital (borrowers and sellers of equity) in exchange for the recipients’ believable claims that they will return amounts of money and credit that are worth more than they were given. While the amount of money in existence is controlled by central banks, the amount of credit in existence can be created out of thin air – i.e. any two willing parties can agree to do a transaction on credit – though this is influenced by central bank policies. In bubbles more credit is created than can be later paid back, which creates busts. When capital contractions occur, economic contractions also occur, i.e. there is not enough money and/or credit spent on goods, services and financial assets. These contractions typically occur for two reasons, which are most commonly known as recessions (which are contractions within a short-term debt cycle) and depressions (which are contractions within deleveragings). Recessions are typically well understood because they happen often and most of us have experienced them, whereas depressions and deleveragings are typically poorly understood because they happen infrequently and are not widely experienced. 1 Credit can be created on the spot between consenting parties. The idea of money going around via "velocity" and adding up to nominal GDP is a misleading description of what happens. 2 State and local governments are of course still significant. 3 A short-term cycle, (which is commonly called the business cycle), arises from a) the rate of growth in spending (i.e. total $ funded by the rates of growth in money and credit) being faster than the rate of growth in the capacity to produce (i.e. total Q) leading to price (P) increases until b) the rate of growth in spending is curtailed by tight money and credit, at which time a recession occurs. In other words, a recession is an economic contraction that is due to a contraction in private sector debt growth arising from tight central bank policy (usually to fight inflation), which ends when the central bank eases. Recessions end when central banks lower interest rates to stimulate demand for goods and services and the credit growth that finances these purchases, because lower interest rates 1) reduce debt service costs; 2) lower monthly payments (de- facto, the costs) of items bought on credit, which stimulates the demand for them; and 3) raise the prices of income-producing assets like stocks, bonds and real estate through the present value effect of discounting their expected cash flows at the lower interest rates, producing a “wealth effect” on spending. In contrast: A long-term debt cycle, arises from debts rising faster than both incomes and money until this can’t continue because debt service costs become excessive, typically because interest rates can’t be reduced any more. A deleveraging is the process of reducing debt burdens (e.g. debt/income ratios). Deleveragings typically end via a mix of 1) debt reduction3, 2) austerity, 3) redistributions of wealth, and 4) debt monetization. A depression is the economic contraction phase of a deleveraging. It occurs because the contraction in private sector debt cannot be rectified by the central bank lowering the cost of money. In depressions, a) a large number of debtors have obligations to deliver more money than they have to meet their obligations, and b) monetary policy is ineffective in reducing debt service costs and stimulating credit growth. Typically, monetary policy is ineffective in stimulating credit growth either because interest rates can’t be lowered (because interest rates are near 0%) to the point of favorably influencing the economics of spending and capital formation (this produces deflationary deleveragings), or because money growth goes into the purchase of inflation hedge assets rather than into credit growth, which produces inflationary deleveragings. Depressions are typically ended by central banks printing money to monetize debt in amounts that offset the deflationary depression effects of debt reductions and austerity. To be clear, while depressions are the contraction phase of a deleveraging, deleveragings (e.g. reducing debt/income ratios) can occur without depressions if they are well managed. (See our “An In-Depth Look at Deleveragings.”) Differences in how governments behave in recessions and deleveragings are good clues that signal which is happening. For example, in deleveragings, central banks typically “print” money that they use to buy large quantities of financial assets in order to compensate for the decline in private sector credit, while these actions are unheard of in recessions.4 Also, in deleveragings, central governments typically spend much, much more to make up for the fall in private sector spending. But let’s not get ahead of ourselves. Since these two types of contractions are just parts of two different types of cycles that are explained more completely in this template, let’s look at the template. 3 Debt reductions take the form of some mix of debt writedowns (so the amount of debt to be repaid is reduced), the timing of debt payments being extended and interest rates being reduced. 4 These show up in changes in their balance sheets that don’t occur in recessions. 4 The Template: The Three Big Forces I believe that three main forces drive most economic activity: 1) trend line productivity growth, 2) the long- term debt cycle and 3) the short-term debt cycle. Figuratively speaking, they look as shown below: 1. 2. 3. What follows is an explanation of all three of these forces and how, by overlaying the archetypical "business" cycle on top of the archetypical long-term debt cycle and overlaying them both on top of the productivity trend line, one can derive a good template for tracking most economic/market movements. While these three forces apply to all countries’ economies, in this study we will look at the U.S. economy over the last hundred years or so as an example to convey the template. In “An In-Depth Look at Deleveragings” and “Why Countries Succeed and Fail Economically” we will look at these cycles across countries. 5 1) Productivity Growth As shown below in chart 1, real per capita GDP has increased at an average rate of a shade less than 2% over the last 100 years and didn’t vary a lot from that. (For an explanation for how and why this varies by country and over time, see the accompanying study “Why Countries Succeed and Fail Economically.”) This is because, over time, knowledge increases, which in turn raises productivity and living standards. As shown in this chart, over the very long run, there is relatively little variation from the trend line. Even the Great Depression in the 1930s looks rather small. As a result, we can be relatively confident that, with time, the economy will get back on track. However, up close, these variations from trend can be enormous. For example, typically in depressions the peak-to-trough declines in real economic activity are around 20%, the destruction of financial wealth is typically more than 50% and equity prices typically decline by around 80%. The losses in financial wealth for those who have it at the beginning of depressions are typically greater than these numbers suggest because there is also a tremendous shifting of who has wealth. Chart 1 Real GDP Per Capita (2008 Dollars, ln) 1.0 1.5 2.0 2.5 3.0 3.5 4.0 00 10 20 30 40 50 60 70 80 90 00 2.8% 0.8% 1.8% 0.2% 4.1% 2.1% 3.0% 2.3% 2.0% 2.1% 1.1% so far Sources: Global Financial Data & BW Estimates Swings around this trend are not primarily due to expansions and contractions in knowledge. For example, the Great Depression didn't occur because people forgot how to efficiently produce, and it wasn't set off by war or drought. All the elements that make the economy buzz were there, yet it stagnated. So why didn't the idle factories simply hire the unemployed to utilize the abundant resources in order to produce prosperity? These cycles are not due to events beyond our control, e.g., natural disasters. They are due to human nature and the way the credit system works. Most importantly, major swings around the trend are due to expansions and contractions in credit – i.e., credit cycles, most importantly 1) a long-term (typically 50 to 75 years) debt cycle (i.e., the “long wave cycle”) and 2) a shorter-term (typically 5 to 8 years) debt cycle (i.e., the “business/market cycle”). About Debt Cycles We find that whenever we start talking about cycles, particularly the "long-wave” variety, we raise eyebrows and elicit reactions similar to those we’d expect if we were talking about astrology. For this reason, before we begin explaining these two debt cycles we'd like to say a few things about cycles in general. A cycle is nothing more than a logical sequence of events leading to a repetitious pattern. In a capitalist economy, cycles of expansions in credit and contractions in credit drive economic cycles and they occur for perfectly logical reasons. Each sequence is not pre-destined to repeat in exactly the same way nor to take 6 exactly the same amount of time, though the patterns are similar, for logical reasons. For example, if you understand the game of Monopoly®, you can pretty well understand credit and economic cycles. Early in the game of Monopoly®, people have a lot of cash and few hotels, and it pays to convert cash into hotels. Those who have more hotels make more money. Seeing this, people tend to convert as much cash as possible into property in order to profit from making other players give them cash. So as the game progresses, more hotels are acquired, which creates more need for cash (to pay the bills of landing on someone else’s property with lots of hotels on it) at the same time as many folks have run down their cash to buy hotels. When they are caught needing cash, they are forced to sell their hotels at discounted prices. So early in the game, “property is king” and later in the game, “cash is king.” Those who are best at playing the game understand how to hold the right mix of property and cash, as this right mix changes. Now, let’s imagine how this Monopoly® game would work if we changed the role of the bank so that it could make loans and take deposits. Players would then be able to borrow money to buy hotels and, rather than holding their cash idly, they would deposit it at the bank to earn interest, which would provide the bank with more money to lend. Let’s also imagine that players in this game could buy and sell properties from each other giving each other credit (i.e., promises to give money and at a later date). If Monopoly® were played this way, it would provide an almost perfect model for the way our economy operates. There would be more spending on hotels (that would be financed with promises to deliver money at a later date). The amount owed would quickly grow to multiples of the amount of money in existence, hotel prices would be higher, and the cash shortage for the debtors who hold hotels would become greater down the road. So, the cycles would become more pronounced. The bank and those who saved by depositing their money in it would also get into trouble when the inability to come up with needed cash caused withdrawals from the bank at the same time as debtors couldn’t come up with cash to pay the bank. Basically, economic and credit cycles work this way. We are now going to look at how credit cycles – both long-term debt cycles and “business cycles” – drive economic cycles. How the System Works Prosperity exists when the economy is operating at a high level of capacity: in other words, when demand is pressing up against a pre-existing level of capacity. At such times, business profits are good and unemployment is low. The longer these conditions persist, the more capacity will be increased, typically financed by credit growth. Declining demand creates a condition of low capacity utilization; as a result, business profits are bad and unemployment is high. The longer these conditions exist, the more cost-cutting (i.e., restructuring) will occur, typically including debt and equity write-downs. Therefore, prosperity equals high demand, and in our credit-based economy, strong demand equals strong real credit growth; conversely, deleveraging equals low demand, and hence lower and falling real credit growth. Contrary to now-popular thinking, recessions and depressions do not develop because of productivity (i.e., inabilities to produce efficiently); they develop from declines in demand, typically due to a fall-off in credit creation. Since changes in demand precede changes in capacity in determining the direction of the economy, one would think that prosperity would be easy to achieve simply through pursuing policies which would steadily increase demand. When the economy is plagued by low capacity utilization, depressed business profitability and high unemployment, why doesn't the government simply give it a good shot of whatever it takes to stimulate demand in order to produce a far more pleasant environment of high capacity utilization, fat profits and l