Classical theory in economy has been that all market clear. Before 2nd World War economic policies about markets relied mostly on Say's law: economy will reach to point of no unemployment and is then in equilibrium. This view is called the general equilibrium theory. So no intervention to market forces was concerned to be necessary.
But between 2 world wars it was seen through the economic crisis, that this law is not working properly. Either the markets reached to equilibrium too slowly or were unstable by their nature, the economy went through trading cycles and unemployment was usually high.
Then a man called Keynes came up with his own views of market equilibrium. He thought it is determined by the aggregate supply and demand, which don't necessarily work in the same way. It follows that if there is too much supply of labour and too little demand for it then economy can be in equilibrium with unemployment. So can the equilibrium occur with unemployment.
Keynes had a model of the economy consisting of firms and households as shown in the diagram below:
This is called the circular flow (money is black lines and factors of productions move the other way). Withdrawal (W) is any part of the income which is not passed on within the circular flow and injection(J) is an addition to the circular flow of income which does not come from the expenditure of domestic households.
This model assumes that everybody's expenditure is somebody else's income. This model of the economy assumes that there is a potential national income (YF) which is not fully reached (spare capacity in economy). This potential can actually be exceeded if everybody would work overtime etc. So it is also assumed, that there is some unemployment.
It is assumed that there is no inflation which would just complicate the model. No change in technology is also assumed.
In that model Y = GNP, because the money coming from all the things produced and sold will eventually be paid to households that do the work.
Withdrawals can be grouped as savings(S)- income not spend by either households or firms, taxes (T) - that is money paid to government and imports - that is money spent on goods bought from abroad.
Injection can be classified as investment(I)-any addition to the real capital stock of the nation or money spent on capital goods and stocks, government spending(G) - includes capital investment and current account balances, but does not include transfer payments (unemployment benefits etc.) and exports - goods / services sold abroad, the amount is determined by the exchange rate, quality and reputation of goods.
The average propensity to consume (APC) is the proportion of income devoted to consumption and is equal to . In the same way there is also average propensity to save (APS), average rate of tax (ART) and average propensity to import and export. Their sum must equal to 1 and they show how much of the income is spent on specific actions.
The marginal propensity to consume (MPC) is the proportion of any addition to income that is devoted to consumption. MPC=. It is assumed to remains constant as Y rises, whereas in real life it tends to fall (more is saved if income is bigger). The same way the marginal propensity to save (MPS) is the proportion of any addition to income that is devoted to savings =. It can be negative in very low incomes as "people eat their savings". Marginal rate of tax (MRT) is the tax one has to pay at a certain addition of income. It was used to be over 1 in specific circumstances (nuisance). Similarly marginal propensity to import and exports (MPI, MPE) are also used. The sum of all propensities is 1 and I will use them later to calculate the multiplier. Main determinants of MPC are income, the distribution of income(lower income groups have higher MPC), consumers' expectations of inflation, unemployment and income, cost and availability of credit and wealth and savings from previous time periods.
The savings will usually be placed to banks who then invest it.
A simplification about investment is also made in following explanations - it is counted as constant, because it consists of two parts: autonomous is determined by factors outside economy(so is constant) and induced is applicable only in the case of full employment and arises due to need for invest much more to new capital when small changes in aggregate demand occur. In practice firms can usually work overtime etc. The same assumption is also made on government expenditure and taxes (tax rate is constant). So new improved flow will look like:
For this system to be in equilibrium either W=J that means T+S+I=E+I+G or Y=G+E+I+C, these are very important results in Keynesian analysis. If Y is bigger than G+E+I+C, then next time firms do not have money to pay for wages, whereas if G+E+I+C is bigger than Y, then Y will eventually rise. This is why Keynes suggested to raise government expenditure (so Y=GNP) will rise and that is also why investment and exports are favoured.
The rise in Y in previous example will bring with it a rise in C, so also the rise in Y again. This is known as the multiplier effect. Multiplier is defined as
K== and was developed by Kahn.
This rise in Y is only temporary, for the change to be permanent the investment must be permanent. If the investment is non-autonomous (rises with income) multiplier is bigger.
The modifying factors of the multiplier are personal disposable income - not all investment will go to PDI (some is depreciation) so multiplier is less than predicted, different types of investment will have different multipliers (e.g. low for defence) and there are time lags involved.
From previous arises the paradox of thrift - the more one intentionally saves the less will be their income (and so savings), because savings are withdrawals that have multiplier effect on Y.
We can also look equilibrium at the point where income must equal expenditure by looking at the graph with ideal expenditure (everything is spent, Y=E) and actual spending (dissaving at very low incomes and saving at higher + spending on tax and net imports (X-M, because M is already included in C figure). As investment, govn. spending and exports are regarded as autonomous, the same amount of expenditure occurs at every level of Y. This is shown in graphs below:
The ideal situation would occur if the equilibrium would be at E. In real life anyway the equilibrium can occur either at higher income or at lower. If the equilibrium occurs at higher Y, then the gap between Y=E line and C+I+G+(X-M) is called a deflationary gap, because economy has tendency to deflate, whereas if Yeq occurs at lower Y, then the gap is called inflationary. There is unemployment of resources present in deflationary gap, because of too little aggregate demand. It is this gap that economies have had most troubles with.
Note that the actual gap is much smaller than the distance the real Y is apart from eqm. Y, because of the multiplier effect.