Population Projection of the Hong Kong NEET Aged CHAN Moon-tong, Ph.D. Hong Kong College of Technology August 2014

Transcription

1 Population Projection of the Hong Kong NEET Aged CHAN Moon-tong, Ph.D. Hong Kong College of Technology August Introduction The term 'NEET' refers to a youth who is both unemployed and economically inactive due to individual and other reasons. It is a short form for Not in Employment, Education or Training. The Hong Kong College of Technology (HKCT) has been conducting research aimed at projecting the population of the Hong Kong NEET in the coming years. Such a projection, based on a set of NEET numbers, could reflect more or less the opportunities a Hong Kong NEET could have for personal enhancement either in the established educational hierarchy or in the job market. The NEET numbers, when soaring for several consecutive quarters, may indicate a grimmer prospect on the part of the NEET for their personal enhancement. The Census & Statistics Department (C&SD) of HKSAR provides us regularly with the data generated from the quarterly General Household Surveys (GHS). The cumulative quarterly data till now range over 57 seasons starting from Q1, 2 and ending at Q1, Research Background NEET draws great concern at the time when the birth rate in Hong Kong have been declining for a number of years and thereby taking effect gradually in the sectors of education at all levels, from primary schools, secondary schools, to private post-secondary institutions. They all have to face the problem of students shortage. However, the drop in birth rate is not the sole cause of the drop in student intake in recent years. As found in surveys, the increase of the NEET population has a part to play in causing the drop in overall student intake as well. It is found that more youths choose to stay at home for an exceptionally long time when there are difficulties in getting a place in a preferred educational program. For those who could not find a favorable job, the chance of staying away from any educational program is becoming higher. Like any educational institute, HKCT is interested to know how big the NEET population would be in the coming years before mapping out a plan of student recruitment. That was of the reasons to prompt HKCT to launch a research based on government data ranging over a period from the first quarter of 2 to the present time. 3. Forecasting For a NEET population projection, one may consult a model that relates the past values of a time series. Time series forecasting method is used in this case to build such a model. The method is 1

2 reliable only when there is no interruption in the relevant time series, such as a change of government policy, catastrophes, economic boom or recession. The occurrence of these events may invalidate the forecasts. Adjustment of such data is required for a better forecast to offset interruption. Figure 1 below shows the quarterly totals of NEET aged 15 19, 2 24, and before adjustment. 1 Fig. 1: Quarterly Numbers of NEET (Q1, 2 - Q1, 214) NEET 2-24 NEET NEET 6 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q Source: Q1, 2 Q1, 214, General Household Surveys, C&SD. The above figure presents a small change with respect to the age group of 2-24 and downward trends to the age groups of and It also shows an obvious seasonality of four quarters per cycle (year) for the three age groups. In this case, population forecast is performed via the statistical package SPSS 2, in order to offset the influence of seasonality. As one may recall, there were intervention events too, i.e., SARS in 23 and financial tsunami in 28. These outburst events definitely affected the performance of forecasting. Thus, an indicator variable was introduced into the model with 1 indicating a time point when an event was thought to have an effect, and when it had no effect. The affected quarters identified were Q2 Q4, 23, Q4, 28, and Q1 Q4, 29. All the above statistical measures had been taken into account when modeling and forecasting were made. The observed values, fitted values, and forecasts for age groups 15 19, 2 24, and are illustrated in Table 1 below: Table 1: Observed Values, Fitted Values/Forecasts of NEET* 2

3 Year Quarter Observed Fitted/Forecast Observed Fitted/Forecast Observed Fitted/Forecast

4 Source: Observed data are from Q1, 2 Q1, 214, General Household Surveys, C&SD. *Notes: 1. The data above are rounded to the nearest hundred. 2. The fitted values and forecasts are not seasonally adjusted. Figure 2 below shows the observed values and fitted values / forecasts of NEET aged

5 Number (seasonally unadjusted): Fig. 2: Observed Values, Fitted Values, and Forecasts of NEET # (15-19, Q1, 2 - Q1, 214) NEET Fitted/Forecasts Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q Source: Observed data are from Q1, 2 Q1, 214, General Household Surveys, C&SD. # The fitted values and forecasts are not seasonally adjusted. The observed values, fitted values, and forecasts for NEET aged are shown in Fig. 2 above. We can see that, except occasionally high values in 23 and 28, there is a downward trend in the series. The population of NEET aged has been decreasing from 295 in Q1, 2 to 9,9 in Q1, 214 with a decrease of 66.4%. Given the continuity, the forecast for Q4, 215 will be 79 with a decrease of 72.2% compared with Q1, 2. The seriousness of the condition for the NEET aged in Hong Kong has been gradually declining. 5

6 Number Fig. 3: Observed Values, Fitted Values, and Forecasts of (2-24, Q1, 2 - Q1, 214) 2-24 NEET Fitted/Forecasts 3 1 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q Source: Observed data are from Q1, 2 Q1, 214, General Household Surveys, The fitted values and forecasts are not seasonally adjusted. The time series for 2-24 does not indicate a trend as shown in Fig. 3 above, in spite of the occurrences of the unforeseen events in 23 and 28. In this case, it is not surprising to find out that the forecasts for the subsequent quarters in 214 to 215 are not significantly different from the observed values of the previous quarters. The number of NEET for Q1, 2 and forecast for Q4, 215 are 34,1 and 36,8 respectively, indicating an increase of 7.9%. 6

7 Number 1 Fig. 4: Observed Values, Fitted Values, and Forecasts of NEET & (15-24, Q1, 2 - Q1, 214) NEET Fitted/Forecasts 6 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q Source: Observed data are from Q1, 2 Q1, 214, General Household Surveys, C&SD. & The fitted values and forecasts are not seasonally adjusted. When the two age groups and 2 24 are combined into 15 24, the downward trend effect of dominates the combined series. We can see a slightly downward trend in Fig. 4 above. The number of NEET in Q1, 2 is 63,5, whereas the forecast of Q4, 215 is 42,5. An obvious decrease of 33.1% is observed during the whole period from Q1, 2 to Q4, 214. The overall impression given by the forecasts is as follows: no significant change of the NEET condition in