While a redistributive land reform has been, perhaps, shelved for the distant future (for political or economic reasons), we shall attempt to show that in the absence of land and ladder to reach the poor through such redistributive mechanisms, we could have alternative ways to help them.

In the following paragraphs we shall show how the growth and composition of income of the rural poor changed over time, and what kind of policy changes should be on board. Admittedly, the discussion would center on the functionally landless households since they comprise a sizeable portion of the total landless households in rural areas.

The total household income of a functionally landless household was estimated to be $819 in 2003-04, and after allowing for change in household size, the per capita income comes to $166. This compares with $111 in 1987-88 and depicts a growth rate of 2.5 percent per annum over time. We observe that, in absolute terms and like land-owning households, there has not been any change in income derived from agriculture. But, relatively, the share of income from agriculture now constitutes a little over one-third, compared to about half in earlier periods.

Noticeably, income from rice farming for functionally landless households, roughly 7 percent of the total income, has been growing at 2.3 percent per annum over the entire period under study. This contrasts with land-owning households which witnessed negative growth rate of income from rice farming. The development on this count seems to reinforce our earlier observations on labour use and changes in tenancy markets. It is not surprising that functionally landless households are entering the tenancy market, increasing operated parcels, and receiving more income from rice farming.

The most important agricultural source of income for the poor in rural areas appears to be non-rice crop activities. Income from this account rose about three fold, depicting a growth rate as high as about 7 percent, and the share almost doubled over time. Similar pattern prevailed with regard to non-crop agriculture. But, as some of those activities, like livestock and poultry raising, require homestead land and capital, the share of income from this remained more or less at the same level.

The idea that the poor households in rural areas mostly depend on wage labour in agriculture does not seem to hold true any more. The share of wage income in the total household income depicted a drastic decline -- from about 28 percent to about 13 percent -- in 1999-2000, and to 12 percent in 2003-04. The income from this source dropped at 3 percent per annum over time. This could be called a significant development since labouring on others' land is not only humiliating but also prone to severe risks because of the unpredictability of nature, and the whims of the landlords.

The prominent source of income of the poor is non-agricultural activities. About two-thirds of the income of these households now comes from non-agricultural pursuits, and has grown at 4.1 percent per annum between 1987-88 and 2003-04. In 1999-2000, the share was as high as 68 percent, but has fallen to about 66 percent in recent years. It is not surprising, therefore, that a large portion of the "missing" labour in agriculture has been absorbed in the non-agricultural sector.

Among non-agricultural sources, trade and business hold the key to the household kitty. In fact, trade and business now account for more than one-fifth of the total household income -- and is surging ahead at about 5 per cent per annum -- for the functionally landless groups. Their presence is also more pervasive than the land owning groups who are generally known to be blessed with capital for carrying out business. We presume that, since this group is the "target group" of the micro-finance institutions spread over every nook and corner of the country, the required capital to do business was supplied by NGOs. But the share of these activities declined from 1999-2000. Do we have to assume that the business environment for the poor has deteriorated in recent times?

Another interesting development should not miss our investigative lenses. Remittance income was once the monopoly of the rich households as they had the money and resources to send members abroad. The functionally landless group, quite obviously, lagged way behind, as they had neither the financial nor the human capital to seize upon the opportunities.

But, of late, the situation seems to have turned for them. Remittance income now constitutes about 13 percent of their household income, compared to only 3 percent in earlier periods. The growth rate of income from this source has been rising astoundingly at 12 percent per annum. In fact, as a source of income for functionally landless households in recent years, remittance puts them almost at par with land owning households.

As mentioned before, the per capita income of the functionally landless households peaked at $166 in the most recent period, and depicted a growth rate of 2.5 percent per annum. The growth rate is above the average for all households and land-owning households. But the inequality syndrome should be considered also. The per capita income of the functionally landless households is 60 percent of that of the land-owning, and 75 percent of all, households. Just sixteen years ago, they had about 59 percent and about 71 percent, respectively, of land-owning and all household income. In other words, the income disparity seems to have dwindled over time.

The field level information point to the following policy implications. Although land is an important asset for rural livelihood, in the absence of land we can provide some non-land opportunities for the landless households. First, the tenancy market could be an important source of operated land for those who do not own land. Second, functionally landless households seem to be opting for non-rice crops, and necessary arrangements for good quality seeds, and training on the optimal use of inputs, could also help them.

Third, necessary capital to carry out petty trade and business could help them increase their income from this source. Fourth, provisions for education for these households would help them with human capital formation and, thus, in eking out a respectable share of income from services and remittances.

By and large, the state has a part to play in increasing their access to education, capital, and a good environment for carrying out activities. Let us provide them with a ladder, if not with land.

Abdul Bayes is Professor of Economics, Jahangirnagar University.