back start next


[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [ 79 ] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] [178] [179] [180] [181] [182] [183]


79

LRAS

FIGURE 5.18 The curve

long-run aggregate supply curve and the aggregate demand

to past and expected future inflation is assumed to be more complicated than the simple formulation in (5.31).

A typical modern Keynesian formulation of aggregate supply is

InPt =lnPt-i + 77* + AdnTt-InTfl-hff, A>0, (5.35)

77t = 77*+\anYt-lnYt) + £f, (5.36)

where = InPt - InPf i is inflation. The A(ln Y - InY) term implies that at any time there is an upward-sloping relationship between inflation and output; the relationship is log-linear for simplicity. Equation (5.36) takes no stand concerning whether it is nominal prices or wages, or some combination of the two, that is the source of the incomplete adjustment." The f term captures supply shocks.

"Equation (5.36) can be combined with Case 2 or 3 of Section 5.4 by assuming that the nominal wage is completely flexible and using the assumption in (5.36) in place of the assumption that P equals P. Similarly, one can assume that wage inflation is given by an expression analogous to (5.36) and use that assumption in place of the assumption that H equals W in Case 1 or 4; this implies somewhat more complicated behavior of price inflation, however.



The standard rule of thumb is that for each percentage point that the unemployment rate exceeds the natural rate, inflation falls by one-half percentage pomt per year. And, as we saw in Section 4.1, for each percentage point that exceeds TT, is roughly 2 percent less than F. Thus if each period corresponds to a year, in equation (5.36) is about one-quarter.

22 One could provide similar accounts of the history of inflaUon and unemploymeni m the 1960s and 1970s, with two complications. First, some of the movements m inflation in 1973-1975 and 1978-80 would be attributed to supply shocks stemming from large oil price increases. Second, the account would posit that the natural rate of unemployment was lower m the 1960s than afterward.

The key difference between (5.36) and the earher models of aggregate supply is the * term. Tautologically, tt* is what inflation would be if output is equal to its natural rate and there are no supply shocks, tt* is known as core, or underlying, inflation. Equation (5.36) is referred to as the expectations-augmented Phillips curve-although, as we will see shortly, modern Keynesian theories do not necessarily interpret tt* as expected inflation.

A simple model of tt* that is useful for fixing ideas is that it equals the previous periods actual inflation:

TT* = TTt-1- (5.37)

With this formulation, there is a tradeoff between output and the change in inflation, but no permanent tradeoff between output and inflation. For inflation to be held steady at any level, output must equal the natural rate. And any level of inflation is sustainable. But for inflation to fall, there must be a period when output is below the natural rate.

This model is much more successful than models with a permanent output-inflation tradeoff at fitting the macroeconomic history of the United States over the past quarter-century. Consider, for example, the behavior of unemployment and inflation since 1980 shown in Figure 5.17. The model attributes the combination of high inflation and high unemployment in the early 1980s to contractionary shifts in aggregate demand with inflation starting from a high level. The high unemployment was associated with falls m inflation (and with larger falls when unemployment was higher), just as the model predicts. Once unemployment feU below the 6-7% range in the mid-1980s, inflation began to creep up. When unemployment returned to this range at the end of the decade, inflation held steady. Inflation again declined when unemployment rose above 7% in 1992, and it again held steady when unemployment fell below 7% in 1993 and 1994. All of these movements are consistent with the model.

Once core inflation is added to the model, it is more convenient to describe the behavior of the economy in output-inflation space than in output-price level space. The aggregate supply curve, (5.36), implies an upward-sloping relationship between output and inflation. And the aggregate demand side of the model implies a downward-sloping relationship between the two variables. To see this, note that for a given value of the



FIGURE 5.19 The AS and AD curves in output-inflation space

previous periods price level, the price level in the current period is an increasing fiinction of the inflation rate. Thus a higher value of inflation implies a lower level of M IP, and hence a lower level of output. These AS and AD curves are shown in Figure 5.19.

Although the model of core inflation in (5.37) is often useful, it has important limitations. For example, if we interpret a period as being fairly short (such as a quarter), core inflation is likely to take more than one period to respond fully to changes in actual inflation. In this case, it is reasonable to replace the right-hand side of (5.37) with a weighted average of inflation over the previous several periods.

Perhaps the most important drawback of the model of aggregate supply in (5.36)-(5.37) is that it assumes that the behavior of core inflation is independent of the economic environment. For example, if the formulation in (5.37) always held, there would be a permanent tradeoff between output and the change in inflation. That is, equations (5.36) and (5.37) imply that if policymakers are willing to accept ever-increasing inflation, they can push output permanently above its natural rate. But the same arguments that Friedman and Phelps make against a permanent output-inflation tradeoff imply that if policymakers attempt to pursue this strategy, workers and firms will eventually stop following (5.36)-(5.37) and will adjust their behavior to account for the increases in inflation they know are going to occur; as a result, output will return to its natural rate.



[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [ 79 ] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] [178] [179] [180] [181] [182] [183]